close

Вход

Забыли?

вход по аккаунту

?

Optical emission spectroscopy investigation of microwave plasmas

код для вставкиСкачать
INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. UMI films the
text directly from the original or copy submitted.
Thus, some thesis and
dissertation copies are in typewriter face, while others may be from any type of
computer printer.
The quality of this reproduction is dependent upon the quality of the copy
submitted.
Broken or indistinct print, colored or poor quality illustrations and
photographs, print bleedthrough, substandard margins, and improper alignment
can adversely affect reproduction.
In the unlikely event that the author did not send UMI a complete manuscript and
there are missing pages, these will be noted. Also, if unauthorized copyright
material had to be removed, a note will indicate the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning
the original, beginning at the upper left-hand comer and continuing from left to
right in equal sections with small overlaps. Each original is also photographed in
one exposure and is included in reduced form at the back of the book.
Photographs included in the original manuscript have been reproduced
xerographically in this copy. Higher quality 6” x 9” black and white photographic
prints are available for any photographs or illustrations appearing in this copy for
an additional charge. Contact UMI directly to order.
Bell & Howell Information and Learning
300 North Zeeb Road, Ann Arbor, Ml 48106-1346 USA
800-521-0600
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
OPTICAL EMISSION SPECTROSCOPY INVESTIGATION OF
MICROWAVE PLASMAS
By
Jayakum aran Sivagnanam e
A TH ESIS
Subm itted to
M ichigan State University
in partial fulfillm ent o f the requirements
for the degree of
M A STE R O F SCIEN CE
D epartm ent o f Electrical and C om puter Engineering
1998
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UMI Number: 1395450
UMI Microform 1395450
Copyright 1999, by UMI Company. All rights reserved.
This microform edition is protected against unauthorized
copying under Title 17, United States Code.
UMI
300 North Zeeb Road
Ann Arbor, MI 48103
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ABSTRACT
O P T IC A L E M IS S IO N S P E C T R O S C O P Y IN V E S T IG A T IO N O F M ICRO W A V E
PL A SM A S
By
Jayakum aran Sivagnanam e
M icrow ave cavity plasm a reactors are being used fo r a range o f materials process­
ing applications including the deposition o f diam ond thin film s and the etching/surface
treatm ent o f sem iconductor m aterials during device fabrication. For all these microwave
plasm a processes a concise understanding o f the plasm a species concentration and ener­
gies are often still lacking. T he plasm a param eters like gas tem perature and species con­
centration, H, C 2 and CH, for som e o f the com m only used plasm as for diam ond thin film
deposition, nam ely H 2 - C H 4 and A r - H 2 - C H 4 were analyzed using optical emission
spectroscopy.
The addition o f sm all am ounts o f N 2 has been show n to affect the deposition rate
and characteristics o f the diam ond film . H ence the effect o f nitrogen on these plasmas was
also studied. W ith the 0.1 - 1% addition o f nitrogen no change in the gas temperature was
observed. H ow ever an increase in the am ount o f atomic hydrogen was observed with
nitrogen concentration.
The density o f high energy electrons in a com pact ion source was also analyzed.
This was achieved by observing the doubly ionized argon em ission lines. The results indi­
cate the presence o f high energy electrons (greater than 27eV ) in the plasm a o f a com pact
ion source.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Dedicated to
Almighty
iii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ACKNOWLEDGEMENTS
I w ould like to express my profound gratitude to Dr. Tim othy G rotjohn. It was his
constant encouragem ent, support and untiring help that led to the com pletion o f this
project. H e has given his knowledge, his tim e and his support to guide m e through the
w ork presented here. I would also like to thank Dr. Jes A sm ussen and Dr. Tim Hogan for
their constructive remarks on the project and for being a part o f the exam ining com m ittee.
Thanks are due to Dr. Anatoly Vikharev, Alexander(Sasha) Kolysko and D m itry Radischev o f the Russian Academy o f Sciences for their contribution to the project, especially
the gas tem perature measurements. I wish to express my gratitude to m y parents and fam­
ily m em bers for their love and support. I w ould also like to thank Bo Keu Kim, M eng-H ua
Tsai, M ark Perrin, A m ir Khan and W en-Shin Huang for their advice and friendship.
iv
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE OF CONTENTS
List of Tables
iii
List of Figures
ix
1 Introduction
1
1.0 M o tiv a tio n ................................................................................................................................... 1
1.2 O b je ctiv es....................................................................................................................................2
1.3 T hesis O u tlin e.............................................................................................................................3
2 Equipment and Experimental Method
4
2.1 Introduction to equipm ent u sed ............................................................................................. 4
2.2 M ultipolar E C R plasm a reactor sy stem ...............................................................................4
2.2.1 D escription o f the Plasm a Source - the M PD R 6 1 0 .............................................4
2.2.2 T h e m icrow ave ap p aratu s............................................................................................ 7
2.2.3 G as/v acu u m system s..................................................................................................... 8
2.3 H igh p o w er resonant cavity m icrow ave reacto r..............................................................10
2.4 S pectrom eter- system # 1 .......................................................................................................12
2.5 D ata acquisition and signal processing for spectrom eter system # 1 ..........................13
2.6 D iode array d etecto r - spectrom eter system # 2 ............................................................... 14
2.7 C ollection O p tic s .....................................................................................................................16
2.7.1 A rran g em en t # 1 ........................................................................................................... 17
2.7.2 A rrangem ent # 2 ........................................................................................................... 17
2.7.3 A rrangem ent # 3 ........................................................................................................... 18
v
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 Identification and analysis of Argon Doubly-ionized atoms
19
3.1 In tro d u ctio n .............................................................................................................................. 19
3.2 T heory o f the diagnostic te c h n iq u e ....................................................................................19
3.4 E xperim ental m e th o d ............................................................................................................ 22
3.3 O bservation and m easurem ent o f the doubly ionized lin es......................................... 23
3.5 R esults and discussion........................................................................................................... 26
3.6 C onclusion ................................................................................................................................33
4 Gas Kinetic Temperature of H2 - CH4 Microwave Plasma
34
4.1 In tro d u ctio n ..............................................................................................................................34
4.2 G as kinetic tem perature m easurem ent th e o ry .................................................................34
4.2.1 H und’s Coupling C ases.............................................................................................. 38
4.3 Experim ental setu p ................................................................................................................. 40
4.4 H ydrogen rotational tem perature........................................................................................ 41
4.4.1 Estim ation o f the rotational tem p eratu re............................................................... 41
4.5 R otational tem perature re su lts............................................................................................. 43
4.6 C o nclusion................................................................................................................................ 50
5 Study of H2 - CH4 - N2 Microwave Plasma
51
5.1 In tro d u ctio n ..............................................................................................................................51
5.2 Experim ental S e tu p ................................................................................................................ 51
5.3 R esults and discu ssio n s......................................................................................................... 52
5.4 C onclusion ................................................................................................................................ 60
vi
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6 Study of Ar - H2 - CH4 - N2 Microwave Plasma
61
6.1 Introduction.............................................................................................................................61
6.2 Experim ental s e tu p ................................................................................................................ 61
6.3 Results and discussion.......................................................................................................... 62
6.5 C onclusion............................................................................................................................. 78
7 Summary of results
79
7.1 C onclusion............................................................................................................................... 79
7.1.1 Argon D oubly-ionized atoms m easurem ents.......................................................79
7.1.2 Gas K inetic T em perature o f H 2 - CH 4 M icrow ave P la sm a ............................. 80
7.1.3 Study o f Ho - CH 4 - N 2 M icrow ave P la s m a ........................................................ 80
7.1.4 Study o f A r-H 2 - C H 4 - N2 M icrow ave P la sm a .................................................. 81
7.2 Recom m endations for future w o rk .................................................................................... 81
Appendix A
84
Appendix B
86
Appendix C
91
References
97
vii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LIST OF TABLES
4 Gas Kinetic Temperature of H2 - CH4 Microwave Plasma
34
Table 4.1: Energy level for the R-branch rotational l i n e s ........................................................ 43
Appendix A
84
Table A: Identification and analysis o f argon doubly ionized a to m s ......................................84
Appendix B
86
Table B: M easurem ents o f A r - H 2 - C H 4 - N 2 M icrowave P la s m a ....................................... 8 6
viii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LIST OF FIGURES
Figure 2.1 M PD R 610 Plasm a S o u rc e .............................................................................................. 6
Figure 2.2 Top view o f the c h am b e r................................................................................................. 7
Figure 2.3 V acuum system o f the plasm a s o u rc e .......................................................................... 9
Figure 2.4 H igh P ow er M icrow ave Plasma R eacto r....................................................................10
Figure 2.5 H igh P ow er M icrow ave Plasma R eacto r..................................................................11
Figure 2.6 Sketch o f the sp ec tro m e te r............................................................................................ 13
Figure 2.7 Sketch o f the diode array d etecto r................................................................................15
Figure 2.8 O ptical set up for light collection - arrangem ent # 1 .............................................. 17
Figure 2.9 O ptical set up for light collection - arrangem ent # 2 .............................................. 18
Figure 2.10 O ptical set up for light collection - arrangem ent # 3 ............................................. 18
Figure 3.1 A tom ic states o f the A r ato m ........................................................................................ 20
Figure 3.2 E xperim ental Set-up for the identification o f Ar*-*- lin e s ......................................23
Figure 3.3 Em ission spectra o f krypton plasm a. Pressure: 3 mT, Flow rate: Isccm , Input
Power: 90 W ...........................................................................................................................................24
Figure 3.4 E m ission spectra o f argon plasma. Pressure: 0.75 m T, F low rate: lsccm , Input
Power: 100 W ........................................................................................................................................ 25
Figure 3.5 V ariation o f A r++ density with pressure for argon plasm a. Flow rate: lsccm .
Input pow er: 40 W ................................................................................................................................29
Figure 3.6 V ariation o f Ar++ density with pressure fo r argon plasm a. Flow rate: lsccm .
Input pow er: 100 W ............................................................................................................................. 30
Figure 3.7 V ariation o f A r++ density with input p o w er for argon plasm a. Flow rate: 1 seem .
Pressure: 0.9 m T .................................................................................................................................. 31
Figure 3.8 V ariation o f A r++ density with flow rate o f A r fo r argon plasm a. Pressure: 0.9
mT. Input pow er: 40 W .......................................................................................................................32
IX
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 4.1 D iagram o f the O bserved Electronic States o f the FL> M o lecu le.........................36
Figure 4.2 Energy Level D iagram for a Band with P and R B ran ch es.................................37
Figure 4.3 Experim ental S et-up for the m easurem ent o f rotational tem perature o f H2.... 40
Figure 4 .4 Em ission S pectra o f the R-Branch rotational lines o f H 2 .................................... 42
Figure 4.5 Boltzm ann p lo t fo r the lines R0 and R 5 -R 1 0 ......................................................... 42
Figure 4.6 V ariation o f rotational tem perature o f H 2 w ith pressure for H 2 plasm a. Flow rate:
H2-200 seem . Input pow er: 400 W ................................................................................................ 45
Figure 4.7 V ariation o f rotational tem perature o f H 2 w ith pressure fo r H 2 -CH 4 plasm a.
Flow rate: H2-200 seem , C H 4-4 seem . Input power: 400 W ....................................................46
Figure 4.8 V ariation o f rotational tem perature o f H 2 w ith input pow er for H 2 -CH 4 plasm a.
Flow rate: H2-200 seem , C H 4-4 seem . Pressure: 30 T o r r ........................................................ 47
Figure 4.9 Variation o f rotational tem perature o f H 2 w ith flow rate for H 2 plasm a. Pressure:
30 Torr. Input power: 4 00 W ........................................................................................................... 48
Figure 4.10 V ariation o f rotational tem perature o f H 2 w ith flow rate o f N2 for H 2 -CH 4 -N2
plasm a. Flow rate: H2-200 seem , C H 4 -4sccm. Pressure: 30 Torr. Input power: 400 W .. 49
Figure 5.1 Variation o f H a line intensity with N 2 concentration. F low rates: CH4-4 seem ,
H2-200 seem. Pressure: 30 T orr. Input Power: 0.8 k W ............................................................. 54
Figure 5.2 V ariation o f H p line intensity with N 2 concentration. Flow rates: CH4-4 seem ,
H2-200 seem. Pressure: 30 T orr. Input Power: 0.8 k W .............................................................55
Figure 5.3 Variation o f H p/H a ratio with N 2 concentration. Flow rates: CH 4-4 seem, H 2200 seem. Pressure: 30 T orr. Input Power: 0.8 k W .................................................................... 56
Figure 5.4 Em ission spectrum o f H 2 -CH 4 -N 2 plasm a. Flow rates: C H 4 -7.2 seem, H2-144
seem, N 2 -0.5 seem. Pressure: 30 Torr. Input Power: 0.8 k W ...................................................57
Figure 5.5 V ariation of C H line intensity with N 2 concentration. Flow rates: CH 4 -7.2 seem .
H2-144 seem. Pressure: 30 T orr. Input Power. 0.8 k W ............................................................. 58
Figure 5.6 V ariation o f C N line intensity with N-> concentration. Flow rates: CH 4 -7.2 seem ,
H2-144 seem. Pressure: 30 T orr. Input Power: 0.8 k W ............................................................. 59
Figure 6.1 Em ission spectrum . Flow rates: Ar- 600 seem , CH4- 3 seem , H 2- 12sccm, N 2- 3
seem. Pressure: 120 T orr. Input Power: 1.09 k W ...................................................................... 64
x
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 6.2 Em ission spectrum . Flow rates: Ar- 100 seem, C H 4- 8 seem, H2- 300sccm, N20 seem . Pressure: 120 Torr. Input Power: 1.621 k W ............................................................... 65
Figure 6.3 Em ission spectrum . Flow rates: Ar- 100 seem , C H 4- 8 seem, H2- 300sccm, N20 seem. Pressure: 120 Torr. Input Power: 1.621 k W ............................................................... 6 6
Figure 6.4 V ariation o f species concentration with H 2 flow rate. Flow rates: Ar- 600 seem,
CH4- 8 seem . Pressure: 120 Torr. Input Power: - 1 . 2 k W ...................................................... 67
Figure 6.4a Variation o f species concentration with H 2 flow rate. Flow rates: Ar- 600 seem,
CH4- 8 seem . Pressure: 120 Torr. Input Power: - 1.2 k W ...................................................... 6 8
Figure 6.5 V ariation o f species concentration with CH 4 flow rate. Flow rates: Ar- 600 seem,
H2- 0 seem . Pressure: 120 T orr. Input Power: - 1.2 k W ..........................................................69
Figure 6.5a Variation o f species concentration with C H 4 flow rate. Flow rates: Ar- 600
seem, H 2- 0 seem. Pressure: 120 Torr. Input Power: ~ 1.2 k W ..............................................70
Figure 6 . 6 V ariation o f species concentration with N 2 flow. Flow rates: A r- 600 seem, H212 seem, C H 4- 8 seem. Pressure: 120 Torr. Input Power: - 1.1 k W .....................................71
Figure 6.7 V ariation o f species concentration with A r flow rate. Flow rates: H 2 - 50-300
seem, CH 4- 8 seem. Pressure: 120 Torr. Input Power: - 1.4 k W ...........................................72
Figure 6.7a V ariation o f species concentration with A r flow rate. Flow rates: H 2 - 50-300
seem, CH 4- 8 seem. Pressure: 120 Torr. Input Power: - 1.4 k W ...........................................73
Figure 6 . 8 V ariation o f C 2 /H a ratio with H 2 flow rate. Flow rates: A r - 600 seem, CH4- 8
seem. Pressure: 120 Torr. Input Pow er: - 1.2 k W ..................................................................... 74
Figure 6.9 V ariation o f C 2 /H a ratio with CH 4 flow rate. Flow rates: A r - 600 seem, H2- 0
seem. Pressure: 120 Torr. Input Pow er: - 1.2 k W ..................................................................... 75
Figure 6.10 Variation o f H p/H a ratio with H 2 flow rate. Flow rates: A r - 600 seem, CH4- 8
seem. Pressure: 120 Torr. Input Pow er: - 1.2 k W ..................................................................... 76
Figure 6 .11 Variation o f H p/H a ratio with H 2 flow rate. Flow rates: A r - 600 seem, H2- 0
seem. Pressure: 120 Torr. Input Pow er: - 1.2 k W ..................................................................... 77
xi
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Chapter 1
Introduction
1.0 Motivation
M icrow ave cavity plasm a reactors are being used for a range of materials process­
ing applications including the deposition o f diam ond thin films and the etching/surface
treatm ent o f sem iconductor m aterials during device fabrication. T he etching applications
are carried out in electron cyclotron resonance (ECR) m icrow ave cavity reactors which
typically operated at low pressure below 10 mTorr. T he deposition o f diamond thin films
use the microwave cavity reactor w ithout the static m agnetic fields and the pressure used
are m uch higher in the range o f 10-120 Torr. For all these m icrow ave plasm a processes a
concise understanding o f the plasm a species concentration and energies are often still
lacking.
For exam ple, during the diam ond deposition process the addition o f small amounts
(1
0 - 1 0 0
ppm ) o f nitrogen can significantly change the deposit film ’s properties and growth
rate [1]. T he reason for the im portance o f nitrogen in the deposition process is not under­
stood. One possible explanation is that the bulk plasm a species concentration or plasm a
tem perature are changed by the nitrogen and that these bulk changes effect the deposition
process. H ence a study o f the various species concentration in the diam ond CVD plasm a
with the addition o f nitrogen w ould be helpful in understanding the role o f nitrogen and
how it affects the plasm a param eters.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
A second exam ple w here additional understanding o f a plasm a process w ould be
useful is the deposition o f nanocrystalline diam ond using A r-C H ^H ^ gas mixture. F o r the
m icrocrystalline and nanocrystalline diam ond grow th, the ratio o f argon to hydrogen has
been found to be a dom inant factor [2]. Therefore an analysis o f the species in the plasm a
m ay shed som e light on their role in the plasm a deposition process.
A third exam ple for additional understanding o f plasm as being o f interest th at is
investigated in this study is the electron energy distribution function in low pressure EC R
plasm as. Specifically, in a com pact EC R ion source used for generating plasmas in m olec­
u lar beam epitaxy m achines, the excitation m echanism in the source is not fully u n d er­
stood. Specifically the density o f high energy electrons in the excitation region needs
investigation.
1.2 Objectives
The m ain objective o f this work is to investigate microwave plasm a discharges
using optical em ission spectroscopy(O ES) in o rd er to add to the understanding o f these
discharges. T he specific objectives are:
1. To study the basic A r plasm a and m easure the am ount o f doubly ionized atoms and
hence high energy electrons present under various process conditions for plasm as cre­
ated in a com pact EC R ion source.
2. To understand the variation o f the plasm a param eters like gas tem perature and species
concentration under various process conditions for plasm as used in diamond C V D
deposition. Specific issues to be studied include the gas tem perature o f diamond CV D
o
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
plasm a, the influx o f nitrogen gas addition on CH 4 -H 2 diam ond deposition plasm as,
and the species concentration variation in CH 4 -AP-H9 plasm as used for nanocrystalline
diam ond deposition.
1.3 Thesis Outline
The main part o f the thesis has been divided into four chapters (C hapters 3-6) with
one chapter devoted to each of the plasm as m entioned in the objectives.
C hapter 2 discusses the equipm ent and the plasm a reactors used in the experi­
m ents. The relevant details needed to understand the experim ental set up has been pro­
vided w ith careful attention to each instrum ent.
C hapter 3 explains the argon doubly ionized atom m easurem ents carried ou t in the
microwave EC R plasm a source. T he results are presented along w ith the discussions.
C hapter 4 details the gas tem perature measurem ents perform ed on the H 2 -C H 4 plasm a. An
insight has been provided to the theoretical aspects involved. T he observations are pre­
sented along with the interpretations. Chapter 5 elaborates the role o f N 2 in the deposition
process as a result of the study o f the H 2 -CH 4 -N 2 plasmas. The influence o f N 2 on the
hydrogen, C H and CN species concentration is given im portance. C hapter
6
presents the
study o f the Ar-H 2 -CH 2 -N 2 plasm as and the results obtained. T he variation o f the constit­
uent species concentration with the process parameters is discussed. C hapter 7 provides
the sum m ary o f the research and discusses future research w ork in this area.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Chapter 2
Equipment and Experimental Method
2.1 Introduction to equipment used
This chapter describes the features o f the resonant cavity microwave discharge
apparatus, especially those features relevant to the optical em ission spectroscopy experi­
m ents. It includes descriptions o f the m icrow ave plasm a reactor, spectrometer, data acqui­
sition system and the collection optics used in the experim ents.
The experim ents w ere perform ed on two different types o f the resonant cavities.
T he diam ond C V D reactor m easurem ents discussed in chapters 4, 5 and
6
were perform ed
in a resonant cavity suited for high pow er and m oderate pressure. The A r++ density m ea­
surem ents discussed in chapter 3 w ere perform ed in the m icrow ave E C R plasm a source
[M PDR 610] suited for low pow er and very low pressure plasm a formation. A ccordingly,
the set-up is described in two sections due to the different configurations o f the plasm a
reactor.
2.2 Multipolar ECR plasma reactor system
2.2.1 Description of the Plasma Source - the MPDR 610
The M icrow ave Plasm a D isk R eactor [MPDR] 610 is a com pact coaxial electron
cyclotron resonance plasm a source. A sketch o f the plasm a source is shown in Figure 2.1.
The cylindrical source is m ade o f stainless steel, and it has an outer diam eter o f about 5.8
cm w ith an application specific overall length [3]. The vacuum seal at one end is m ade by
4
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
a standard 4.5 inch Conflat flange w ith the entire length o f the source inserted into the vac­
uum cham ber. At the other end (near the discharge), m etal-to-fused quartz vacuum seals
are m ade using specially designed ultra high vacuum com patible ring shaped H elicoflex
spring loaded seals. The discharge located at this end o f the plasm a source has a d iam eter
o f 3.5 cm . T he electrom agnetic cavity excitation region consists o f a coaxial coupling sec­
tion w hich is term inated at one end w ith a sliding short, and at the other end, by the dis­
charge. T h e region betw een the sliding short and the discharge is occupied by a coaxial
w aveguide section and an evanescent circu lar waveguide section. The diam eter o f the
inner co n d u cto r com prising the coaxial section is 1.2 cm. A sm all loop antenna is attached
to the sliding short, and this loop excites the T E M mode in the microwave cavity. B etw een
the term inus o f the center conductor and the discharge is an im pedance m atching circu lar
w aveguide section that has a diam eter w hich is too small to support any electrom agnetic
propagating m odes for the 2.45 G H z excitation frequency. The plasm a is confined in a
quartz discharge cham ber with an inner d iam eter o f 3.5 cm and a height o f 4.7 cm . The
quartz tube is surrounded by three ring shaped axially m agnetized perm anent m agnets
which provide the static magnetic flux density w ithin the plasm a cham ber for plasm a co n ­
finem ent and the 875 G auss field strength
for EC R . The magnets
have an outer d iam eter o f
w
w
4.95 cm , a height o f 1.27 cm, and an inner diam eter o f 4.32 cm . T hey are aligned w ith like
poles facing each other.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Type N
M icrow ave
C o n n ecto r
G as F eed
G as L in e
L oop A n ten n a
P la sm a
Region
Air Cooling
S tan d ard
4.5 inchC onflat
Slid in g
S hort
C en ter
C o n d u cto r
P e rm a n e n t
M ag n ets
Figure 2.1 M PD R 610 Plasm a Source
The quartz walls and the m agnets are cooled by com pressed a ir flowing through
the center conductor and blow ing onto the discharge chamber. The positions o f the center
conductor and the sliding short can be independently adjusted in o rd e r to match the dis­
charge load to the input transm ission line im pedance. The discharge load changes as the
plasm a varies with input power, gas flow/pressure, gas type etc., and appropriate tuning
m inim izes the reflected pow er a n d ensures m axim um pow er transfer to the plasm a load.
The cylindrical stainless steel processing cham ber is 10 inch high and has a diam ­
eter o f 18 inches. It has four ports one each connected to the vacuum pum p and the plasm a
source. The other two ports are used as viewing port for the optical diagnostics. The view­
ing ports and the port connected to the source are
8
inch in diam eter. T he transparent win­
dow o f the view ing port has a d iam eter o f 1.8 inches. The ports are orthogonal as shown in
Figure 2.2.
6
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Air „
Cooling
50 Ohm Coaxial cable
from microwave source
MPDR 610 Tuning
Assembly
Discharge
Chamber
► Input gas
Rare earth
magnets
Processing Chamber
Viewport
To
\
Diffusion Pump
Viewport
Figure 2.2 Top view o f the cham ber
2.2.2 The microwave apparatus
M icrow ave energy is supplied by a 2.45 GHz m icrow ave pow er supply (Raytheon
PG M 10X 1). T he experim ents described in Chapter 3 were perform ed with microwave
pow er ranging from 10 Watts to 100 Watts. This pow er range refers to the pow er absorbed
by the cavity which is found by subtracting the microwave pow er reflected by the cavity
7
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
from the pow er incident to the cavity. The m icrow ave circuit includes a three port circula­
tor and dum m y load to protect the pow er supply and a dual directional coupler fo r sam ­
pling both the reflected and incident power.
A 50 ohm coaxial cable w ith type N connectors and a Teflon dielectric filling
m aterial transfers pow er from the bi-directional couplers to the plasm a source. T he coaxial
structure o f the cable is continued into the source body, term inating in the loop antenna as
shown in Figure 2.1. This loop excites the T E M m ode in the coaxial cavity that stretches
from the end o f the sliding short to the terminus o f the center conductor, beyond w hich
evanescent fields are excited in the circular w aveguide section o f the source.
2.2.3 Gas/vacuum systems
T h e vacuum system connected to the plasm a reactor used for the argon doubly
ionized atom identification and m easurem ent (discussed in chapter 3) can be represented
as show n in the Figure 2.3.
T h e vacuum cham ber is connected to a
6
inch diam eter oil diffusion pum p and
backed b y a m echanical pum p. A gate valve isolates the vacuum cham ber from the diffu­
sion pum p stack. Betw een the gate valve and the diffusion pump is a gate conductance
valve and a w ater cooled cold trap that helps in keeping stray pum p oil away from the vac­
uum cham ber. T he diffusion pum p is filled w ith a hydrocarbon-free oil [Fomblin] to allow
the use o f reactive gases. N orm al cham ber base pressures under optim al pum ping co n d i­
tions w ere about 1 0 ° Torr.
8
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Gate
Ionization gauge tube
Conductance
Gate Valve JTo ionization gauge
Valve
Processing Chamber
Diffusion
Pump
Exhaust
Roughing Valve
Foreline Valve
Circulator
Mass Flow
Controller
Mechanical Pump
Dummy Load
-Gas Tank
Microwave Power
Supply
F igure 2.3 Vacuum system o f the plasm a source
T h e gate valve, roughing valve and the foreline valves are pneum atically co n ­
trolled through a specially designed sw itch box. T he pressure o f the processing ch am b er is
m easured using an ionization gauge an d a capacitance m anom eter (100 mTorr full range),
w hile the foreline and the diffusion pum p stack pressure are m easured using th erm o co u ­
ples.
G as flow into the M PD R is controlled through an MKS 247C four channel read­
out/controller. Pressure conditions inside the vacuum cham ber are controlled by a gate
conductance valve. All com ponents o f this system are made o f U ltra H igh V acuum [UHV]
com patible steel.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2.3 High power resonant cavity microwave reactor
The high pow er microwave cavity plasm a reactor is a resonant cavity microwave
discharge operating in the TM 0 j 3 m ode. T he reactor includes a 7 inch inside diam eter
m icrow ave cavity that can be electrom agnetically tuned with a sliding short and an adjust­
able probe. T he tuning process consists o f m atching the com plex im pedance o f the cavity
(Zin= R in+ jX in) to the transm ission line im pedance which carries the microwave power
source to the cavity. The cavity effectively directs intense microwave energy into the
plasm a source region. A C ober m odel S6F/4503 2.45 G H z microwave pow er supply was
used to provide 0 .2 - 2
.0
kilowatts o f m icrow ave power.
Microwave Input Probe
Sliding Short
Window
Air Cooling
Plasma
Quartz Dome
Substrate Holder
Figure 2.4 High Pow er M icrow ave Plasm a R eactor
10
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The hydrogen discharge experim ents d iscussed in chapter 4 and 5 w ere perform ed
in the sam e reactor as shown in Figure 2.4. T he experim ents discussed in ch ap te r
6
were
carried out in the reactor shown in Figure 2.5. T hese reactors differ in the shape o f the
quartz dom e used and the other basic configuration rem ains the same. All th e experim ents
w ere perform ed at pressures o f 10-120 Torr. T he M KS 247C four channel readout/control­
ler regulates the flow o f hydrogen, m ethane and nitrogen.
Microwave Input Probe
Sliding Short
Window
Air Cooling
Quartz Dome
Substr
Substrate Holder
Figure 2.5 High Power M icrowave Plasm a R eactor
T he source is designed to operate at high pow er and moderate pressure conditions.
H ence the vacuum system is not as com plex as the previous system. The c h am b er is
directly connected to a mechanical pum p. A throttle valve connected betw een the pump
and the cham ber is used for fine control o f the ch am b er pressure. Since high powers are
11
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
involved the quartz dom e gets heated. A ir cooling is provided to the quartz dom e by a duct
with a fan connected to the other end. T h e heated air flows out through the mesh window,
w hich is also used as a view ing port.
2.4 Spectrometer- system #1
T he spectrom eter used for some o f the m easurem ents is a M cPherson M odel
216.5, 0.5 meter, f/8.7, plane grating scanning m onochrom ator. It is designed to operate in
the w avelength from 1050 A to 160,000 A by interchanging the gratings. The 2400
grooves/m m grating can be used in the 1050 A - 5000 A w avelength range and the 1200
grooves/m m grating can be used in the 1050 A -10,000 A w avelength range. The spec­
trom eter has a resolution o f 0.2 A and 0 .4 A for the 2400 line/m m and 1200 lines/m m grat­
ing respectively with input and output slits o f the spectrom eter set at 10 p m wide. The
optical system o f the spectrom eter consists o f two concave m irrors an d a plane grating.
A collim ating m irror is 3” in diam eter and has a 0.5 m eter focal length. A focus­
sing m irror is 6 ” in diam eter and has a 0.5 m eter focal length w ith a 4 ” focal plane. The
m irrors are m ounted in rigid alum inum holders. The input and the output slits are adjust­
able and are provided w ith a opening range o f
1 0
microns to
2
mm.
A photom ultiplier tube is m ounted at the exit slit. T he M cP herson m odel scanning
m onochrom ator houses a E G I-G E N C O M R P IQ L /2 0 photom ultiplier tube. For all the
experim ents a bias voltage o f -800 V was applied to the p hotom ultiplier tube in order to
12
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
get a good signal-to-noise ratio. An Oriel m odel 70705 photom ultiplier p o w er supply was
used for this purpose.
Input slit
Concave mirrors
Output slit
Gratin
Photomultiplier
Figure 2.6 Sketch o f the spectrom eter
T he principle used in the operation o f the m onochrom ator is diffraction. T he grat­
ing in the m onochrom ator diffracts the incom ing light. The angle o f diffraction varies with
the w avelength. H ence at any time only the diffracted signal o f a particular w avelength is
collected by the concave mirror and focussed on to the photom ultiplier. T he w avelength
scan is done by tilting the grating about its vertical axis as shown in Figure 2.6. The scan­
ning m otor has fixed speeds of 0.5, 1, 2, 10, 20, 50, 100, 200, 500, 1000, 2000 A per
m inute. U sually the data acquisition system collects around 4 sam ples/sec. fro m the spec­
trom eter. D epending on the wavelength range and accuracy desired, the sp eed o f the scan­
ning m otor is selected.
13
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2.5 Data acquisition and signal processing for spectrometer system #1
The data acquisition and processing w ere done by a sem i-autom ated system. The
spectrom eter has a fixed speed m otor used fo r scanning, that needs to be operated m anu­
ally. Except for the start-up o f the scanning m otor o f the spectrom eter, the rest o f the pro­
cess was autom ated. The output current from the photom ultiplier tube was read by a
K eithley 485 A utoranging picoam m eter. T he output current was usually in the range o f a
few nanoam peres, depending on the intensity o f the signal. The picoam m eter was con­
nected to a personal com puter by a EEEE-488 interface. The IEEE-488 bus was connected
to a N ational Instruments GPIB card in the com puter. The software consisted o f a Q uick­
BASIC program that was used to co llect the data at the desired interval. The processing
was done using MATLAB. The code for the softw are part is included in A ppendix A.
2.6 Diode array detector - spectrometer system #2
The spectroscopic m easurem ents described in chapter
6
w ere perform ed using the
O R IEL Instaspec diode array detector. A spectrograph/diode array com bination is a
replacem ent for a m otor driven m onochrom ator and photom ultiplier tube. The light source
is reduced to a point or thin line at the en tran ce slit, and the spectral content o f the image is
m easured. The dispersed spectrum is then projected as a continuous band o f w avelengths
onto a diode array placed at the focal plane.
The diode arrays have their own advantages and limitations com pared to the pho­
tom ultiplier tube. The m ost im portant difference between the photom ultiplier tube and
14
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
diode arrays is the m ultichannel nature o f the arrays. The large nu m b er o f photodiodes
constitute a series o f detectors in close proxim ity, and this results in the ability to detect a
num ber o f independent events sim ultaneously. Thus, rather than being a single detector,
the diode array is a one dim ensional array o f detectors. This m ultichannel nature o f the
diode array results in an increased rate o f data acquisition for m ultiple data points and
eliminates the requirem ent to move the im age relative to the d etecto r as in the scanning
monochromator.
Diode Array
Gratin
Concave mirrors
Glass Plate
To Computer
Input Slit
Figure 2.7 Sketch o f the diode array detector
The lim iting spatial resolution o f a diode array is the elem en t spacing along the
detector. The w avelength resolution o f the array is dependent on th e dispersed w avelength
range over the array and the elem ent size. The spectroscopic w avelength resolution is
15
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
dependent on the input slit width, the total bandpass o v er the array and the num ber o f ele­
m ents in the array. T he diode array detection sy stem is fully autom ated and is controlled
by the softw are provided by the manufacturer.
2.7 Collection Optics
T he collection optics usually consisted o f an arrangem ent o f lenses o f different
focal lengths depending on the strength o f the sig n al and the am ount o f coupling desired.
Three different arrangem ents were used for the various experim ents. For simplicity, the
positioning o f the lenses and the selection o f the focal lengths w ere based on the results of
the calculations obtained by considering the em issio n beam from the plasm a source as
straight rays. Fine tuning o f the output was done m anually by viewing the focus o f the out­
put beam on a sheet o f white paper. The fiber o p tic cable o f the spectrom eter was aligned
so that the focussed beam had m axim um coupled to the core o f the fiber. A dditional focus­
sing was not needed since the diam eter o f the co re o f the fiber optic cable was large [1-3
mm]. Since m easurem ents were perform ed on th e w hole plasm a, spatial resolution w as
not required. This avoided the necessity o f co llectin g light from specific parts o f the
plasm a. In cases w here the signal was very w eak, a b lack cloth was used to cover the
entire collection optics set up to the entrance o f the fiber, to prevent interference from the
am bient light. T he optical arrangem ents are p resen ted below in the increasing order o f
com plexity along with the necessary details.
16
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Two types o f fibers w ere used. O ne o f them is an optical cable w ith a diam eter o f 3
m m and the o th er one consisted o f a m ultim ode fiber with a core d iam eter o f
1
mm.
2.7.1 Arrangement #1
T his arrangem ent is very sim ple and requires the direct focu ssin g o f the multim ode fiber to the plasm a. This arrangem ent was used w hen the p lasm a discharge was very
bright and there was enough signal coupled through the m ultim ode fiber.
Plasma Source
Substrate
Substrate holder
Optical Fiber
Figure 2.8 Optical set up for light collection - arrangem ent #1
2.7.2 Arrangement #2
T his arrangem ent involves a single biconvex lens o f d iam eter 5.1 cm and focal
length 5 cm . T he optical cable was used in this set up. T he optical set up is shown in Fig­
ure 2.9. T he distances S and S ’ and the focal le n g th /o f the lens is governed by the relation
I _ 1
f ~ S + S'
17
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Lens
Plasma Source
Substrate
Substrate holder
Optical Fiber
Figure 2.9 O ptical set up for light collection - arrangem ent #2
2.7.3 Arrangement #3
This arrangem ent involves three biconvex lenses, two o f them with diam eter 6.3
cm and the third one with a diam eter o f 5.1 cm . They had focal lengths o f 30 cm, 15 cm
and 5 cm respectively. Since the signal from the plasm a was weak, the collected light was
directly focussed on to the input slit o f the spectrom eter system #1. The arrangem ent is
shown in Figure 2.10.
Source
f=30cm , 0 = 6 .3cm
f=15cm, 0 = 6 .3cm f=5cm, 0 = 5 .lcm
Input Slit
|2 c m
15cm
-20cm
20cm
Figure 2.10 O ptical set up for light collection - arrangem ent #3
18
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Chapter 3
Identification and analysis of Argon Doubly-ionized
atoms
3.1 Introduction
This chapter deals w ith the identification and analysis o f the doubly-ionized Argon
atom s in a microwave E C R plasm a discharge. This experim ent is aimed at identifying the
A r++ density variation and to a lesser extent the electron tem perature variation w ith chang­
ing process param eters in a c om pact EC R ion source. T he m otivation for studying Ar++
em ission is that higher energy electrons in excess o f 27 eV are required to produce A r++
em ission. Hence the observation o f A r++ em ission intensities is a relative in d icato r o f the
num ber o f high energy electrons. T he first section briefly discusses about the optical tech­
nique and the research literature used as a reference in identifying the Ar++ em ission lines.
T he subsequent sections describe the experim ent and the data obtained. T he last section
contains a discussion o f the results.
3.2 Theory of the diagnostic technique
The spectroscopic radiation is em itted w hen a bound electron m akes a transition in
an atom or ion. The observed intensity o f the radiation thus em itted depends on
1.the probability o f there being a bound electron in the upper level o f the transition,
2
.the atomic probability o f the transition in question, and
3.the probability of the photons thus produced escaping from the volume o f the plasm a
w ithout being reabsorbed.
19
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
T he relevant atom ic states o f the A rgon atom can be represented as shown in Fig­
ure 3. L. As it could be seen from the figure, the ionization potential o f the A r atom is 15.76
eV. B y im parting 27.628 eV to the Ar+ ion an electron from the valence shell o f the Ar+
ion can be rem oved resulting in an Ar++ ion. To cause further ionization 40.9 eV is
required.
Ar'. + + + .
40.9 eV
ArH
333.6 nm
24.38 eV
28.1 eV
Ar'.+ + n
T
27.628 eV
A r4
■f
15.76 eV
ArFigure 3.1 A tom ic states o f the A r atom
20
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 3.1 depicts the observed 3336.13 A transition o f the excited Ar++* ion. The
valence electron o f the A r'-1" ion is excited to an interm ediate state by absorbing 28.1 eV
via an electron collision. This energy state is w ell below the ionization potential o f the
valence electron o f the A r++ ion. From this interm ediate state, the electron drops spontane­
ously to a lower interm ediate state 3.72 eV below it, sim ultaneously radiating the excess
energy as a photon. It is this radiation w hich is observed as a 3336.13 A line in the spectra.
Also the energy separation betw een Ar++-Ar+ and A r++*-Ar++ are alm ost the same. This
shows that the intensity o f the 3336.31 A is a good indicator o f the density of the Ar++
ions since both A r++* and A r++ are created by electrons o f sim ilar energy. The doubly-ion­
ized A r em ission lines w ere identified using the data presented by Striganov et al. [4], T he
spectral lines that are presented in this reference have been m easured with an accuracy o f
more than 1 A.
For a better understanding o f the principles involved in the measurement, let us
show that the transition that is considered is indeed indicative o f the doubly ionized atom
density. The ion form ation results due to the excitation o f electrons to higher energy lev­
els. In an atom, the energy transfer required for the excitation o f the bound electrons from
one energy level to the other takes place due to the collision o f the free high energy elec­
trons w ith the bound electrons in the low er energy level. The collision frequency o f the
free electrons at a given energy level is given by
Ver = ne\o (E )f(E )d E
where E is the electron energy, a is the excitation cross section, ne is the electron density
and f(E) is the free electron distribution function. For A r+ to Ax'-*" transition, VeT can be
2 1
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
approxim ated as V ex~ Cine(Eex), w here C / is a constant. H ence basically the collision fre­
quency depends on the num ber o f electrons at or above the given energy level Eex. Sim i­
larly for the A rH' to A r++* transition,
can be approxim ated as V ex~ C2ne(Eex), where
C2 is another constant. Thus the num ber o f ions excited from Ar*-1' to Ar++* state depends
on the same set o f higher energy free electrons that form ed the Ar*4- state. The 333.6 nm
transition that is observed occurs due to the loss o f energy o f the electrons as they move
from the Ar++* state to another A r++* state. The observation o f this transition is thus a
good indicator o f the density o f the electrons with energy above E^ and the density o f
Ar++ ions.
3.4 Experimental method
The experim ents were perform ed using the M PD R 610 microwave plasm a reactor.
The optical em ission from the discharge was focused to a fiber using a lens. The arrange­
ment #2 of the collection optics shown in Figure 2.9 was used. The other end o f the fiber
was focussed to the input slit o f the spectrom eter system #1 shown in Figure 2.6. The
entrance and exit slits o f the spectrom eter w ere 50 p. wide by 2 cm high. The resolution o f
the spectrom eter used was 1A with 50 p. slits. The 0.5 m eter spectrom eter contains a 2400
lines/mm grating. The wavelength range scanned during the experim ents is 3320 A- 3550
A. The optical fiber used in the experim ent allow ed the signal in this wavelength range
without attenuation. T he experim ental setup is as shown in Figure 3.2. The data acquisi­
tion system is the sam e as that discussed in the chapter 2 .
22
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Microwave Power
f=5cm
Processing Chamber
Plasma
Quartz Window
Input Gas
Optical
Fiber >
Monochromator
Photomultiplier
IEEE-488 bus
Computer
Picoammeter
Figure 3.2 E xperim ental Set-up for the identification o f Ar++ lines
3.3 Observation and measurement of the doubly ionized lines
T he intensity o f the doubly ionized atom em ission lines gives more details about
the param eters o f the plasm a. Ionization occurs w hen the high energy particles collide
with an atom o r ion. For double ionization to o ccur the energy o f the charged particles
needs to be particularly high. H ence the intensity o f the doubly ionized lines in the spectra
is a direct representation o f the presence o f high energy electrons in the plasma.
F or the m easurem ents at least 10 doubly ionized lines o f A r were identified in i­
tially. In order to m ake sure that the identified lines belonged to Ax and not any o th er
im purity atom , argon was replaced by krypton and the sam e w avelength range w as
scanned. By com parison, m ore than half the num ber o f peaks w ere identified as im puri-
23
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ties. M ore scans with greater resolution w ere taken in the wavelength range o f the rem ain­
ing peaks to check for their presence. O ne such w avelength scan is show n in Figure 3.3.
Finally three lines were chosen fo r the A r++ m easurem ents. In the m easurem ents, doubly
ionized lines with w avelengths 3336.13 A (4 s‘ 3 D ° [J=3]- 4 p ‘ 3 F[J=4]), 3344.72 A (4 s‘ 3 D °
[J=2]- 4 p ‘ 3F [J=3]) and 3358.49 A (4 s 1 3 D ° [J= I]- 4 p ‘ 3F [J=2]) w ere considered.
-40
3320
3340
3360
W avelength (A)
3380
3400
Figure 3.3 Em ission spectra o f krypton plasm a. Pressure: 3 mT, Flow rate: lsccm ,
Input Power: 90 W.
24
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
140
120
100
GO
r-
3
_d
c3
1
)
O
3344.72 A
3336.13 A
3358,.49 A
3320
3340
3360
3380
3400
Wavelength (A )
Figure 3.4 Em ission spectra o f argon plasm a. Pressure: 0.75 mT, Flow rate:
lsccm , Input Power: 100 W.
LS (R ussel-Saunders coupling schem e) notation is used here to represent the quan­
tum states. In this schem e the quantum state o f an atom is labeled in the following manner:
Nlk 2S+lU
where L is the total orbital angular momentum quantum number, S the total spin quan­
tum number, and J the magnitude o f the total angular momentum (J=L+S). 2S+1 gives
25
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
the multiplicity o f the quantum state. T he nu m b er N denotes the orbit number (in the B ohr
sense), and / the angular momentum states o f the last k active electrons
M easurem ents were made fo r different values o f incident microwave power, pres­
sure and gas flow rates. For greater accuracy, the area under the peak was calculated,
rather than the peak intensity after subtracting the background noise. The em ission spec­
trum observed in the 3320-3400 A range is show n in F igure 3.4. The peaks corresponding
to the doubly ionized atoms are m arked on the figure. F o r calculating the area o f the peak,
the trapezoidal m ethod [In this num erical integration m ethod, the curve is approxim ated
by small trapezoids and the total area under the cu rv e is obtained by adding the areas o f
the individual trapezoids] was used. T he Q B A SIC co d e that was used to carry out these
measurem ents is presented in Appendix C.
3.5 Results and discussion
The results o f the experim ents are presented in this section. The com plete data
obtained for the various conditions are show n in A ppendix A. The various table entries
show changes in A r++* em issions produced by the p lasm a source input param eter changes.
The source param eters that were analyzed included pressure, input power, and gas flow
rate. The pressure w as varied from 0.4 to 3.0 mTorr, th e input pow er from 10 to 100 w atts,
and the flow rate from
1
to
2 0
seem argon.
Figure 3.5 and 3.6 show the variation in A r4-*- em ission versus pressure. T he line
on these plots indicate the average intensity o f the three em ission lines. The higher A r4-*"
emission occur in these figures at the low er pressures. This is consistent with the electron
temperature increasing at low pressure. B asically, the increased electron tem perature have
more electrons w ith 28 eV or more o f energy. It is these electrons that are needed to pro-
26
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
duce Ar++ and A r 1-1'* as show n earlier in Figure 3.1. Figure 3.5 is at a pow er o f 40 watts.
T he A r 1-*"em issions at this pow er are seen to drop by a factor o f
6
w hen the pressure
changes from 0 .4 m Torr to 3 mTorr. Figure 3.6 is at a higher pow er o f 100 w atts. The A rH"
em ission is seen at this pow er to drop by a factor o f about
2
w hen the pressure changes
from 0.4 mTorr to 3 mTorr.
Figure 3.7 show s the influence o f m icrowave pow er on the Aj *++ em ission. The
flow rate was 1 seem and the pressure was 0.9 mTorr for the data show n in this figure. The
A r++ em ission is seen to rise w ith input pow er to about 40-50 w atts and then the em issions
appear to saturate. This im plies that fo r input pow er levels o f 50-100 w atts the A r1-1- den­
sity is not changing significantly.
The last figure shown is Figure 3.8. T he Ar++ em ission is p lo tted versus argon flow
rate at a pow er o f 40 watts and a pressure o f 0.9 mTorr. The A r++ em ission and hence its
density drops at the higher flow rates. The effect is quite substantial w ith the em ission
dropping by an average factor o f
8
w hen the flow rate changes fro m
1
seem to
2 0
seem.
The typical behaviors expected from a plasm a are: 1) As the pressure is reduced
the electron tem perature increases, and 2 ) as the input m icrowave po w er increases the ion
density Ar+ increases. The variation in the argon Ar++ em ission follow s the expected pres­
sure dependence. Specifically, A r++ production requires high electron tem peratures, and
the higher Ar++ em ission at low pressures in Figure 3.5 and 3.6 sup p o rt the presence o f
m ore high energy electrons at low pressure. The data in Figure 3.7 show s that the Ar++
density is increasing with input pow er from 10-50 watts. This is as expected since more
pow er increases A r+ and hence m ore Ar+ can be ionized to A r++. F o r the higher m icro-
27
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
wave power, the A r++ em ission appears to saturate versus m icrowave pow er increases.
This effect is a deviation from the expected behavior and further, future investigation
w ould be needed to clarify the cause o f this effect.
T he d ata o f A r++ em ission versus argon flow rate in Figure 3.8 shows a strong vari­
ation. It can be concluded from the data that at the large flow rates, the A r++ density is
reduced. Two m echanism s could be possible explanations o f the observed data. M echa­
nism
1
is that at high flow rates the residence tim e o f the plasm a particles in the plasm a
source region is reduced. T he reduction is enough that A r++ ions do not have time to be
created and excited. T he second m echanism is that at the low flow rates, strong neutral gas
heating occurs w hich reduces the electron collision frequency. T he reduced electron colli­
sions raises the average electron energy making A r++ density and em ission larger. Further
experim entation such as gas tem perature m easurem ents w ould help to clarify the responsi­
ble m echanism .
28
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Emission Intensity (arb. units)
0
0.5
1
1.5
Pressure (mT)
2
2.5
Figure 3.5 Variation o f Ar++ density with pressure for argon plasm a. Flow rate:
lsccm . Input power: 40 W
29
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3
0 3336.13 A
D 3344.72 A
0 3358.49 A
0.5
1.5
Pressure (mT)
2.5
Figure 3.6 Variation o f A r++ density w ith pressure fo r argon plasm a. Flow rate:
lscc m . Input power: 100 W
30
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Emission Intensity (arb. units)
03336.13 A
n 3344.72 A
0 3358.49 A
10
20
30
40
70
50
60
Input Power (W)
80
90
100
Figure 3.7 V ariation o f Ar*-*- density with input p o w er fo r argon plasm a. Flow rate
1 seem . Pressure: 0.9 mT.
31
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0 3336.13 A
o 3344.72A
o 3358.49 A
Emission Intensity (arb. units)
4.5
3.5
0.5
Flow rate (seem)
Figure 3.8 Variation o f A r 1-1" density w ith flow rate o f A r for argon plasm a. Pres­
sure: 0.9 mT. Input power: 40 W
32
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3.6 Conclusion
The experim ents have shown the dependence o f A r++ em issions on the process
param eters. It confirm s the presence o f A r++ ions in the system . Also the A r++ ion density
has a large dependence on the flow rate and the pressure. B oth low flow rates and low
pressures produce larger Ar*-1- em issions. T he results obtained presents data for helping to
understand the A r++ density variation and high energy electron density variation in the 610
com pact ion source.
33
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Chapter 4
Gas Kinetic Temperature of H 2 - CH 4 Microwave
Plasma
4.1 Introduction
This ch ap ter introduces the experim ental set-up used to carry out the diagnostic
experim ents on H 2 -C H 4 plasm a gas tem perature and then describes the experim ental
results. For both the growth rate and quality o f the diam ond films the optim um tem pera­
ture is betw een 1100 and 1300 K [5]. It is likely that the deposition process is sensitive to
the gas kinetic tem perature. The gas tem perature helps determ ine the concentration o f var­
ious radicals, because many gas-phase reaction rates are strongly dependent on the gas
kinetic tem perature. The first section details the theory o f optical em ission spectroscopy
used in the gas tem perature m easurem ent. T h e subsequent sections describe the specific
details o f the experim ental setup. Finally the results obtained are discussed.
4.2 Gas kinetic temperature measurement theory
O ptical em ission spectroscopy (O ES) is used to m easure the rotational tem perature
o f H 2 neutral species. The energy separations betw een rotational levels in a given vibra­
tional and electronic state are typically sm all com pared w ith the therm al translational
energy. N early all gas kinetic collisions produce a change in the rotational quantum num ­
ber, w hereas collisions producing a change in the vibrational or electronic quantum num ­
bers usually o ccu r much less frequently. C onsequently, the relative rotational population
distribution in a sufficiently long-lived vibrational state has a Boltzm ann distribution and
the rotational tem perature reflects the gas kinetic tem perature [6 ]. T he rotational tem pera-
34
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ture is derived from m easurem ent o f the relative intensities o f rotational lines within a sin­
gle vibrational band. T he relative rotational line intensities / o f a B oltzm ann distribution
are described by [7]
4.1
w here AT is a constant for all lines originating from the sam e electronic and vibrational
level, u is the frequency o f the radiation, Sj -j - is the appropriate HOnl-London factor,
By, is the molecular rotational constant fo r the upper vibrational level, J is the rota­
tional quantum number, h is the Planck’s constant, c is the speed o f light, k is the Boltz­
mann’s constant and Tr is the rotational temperature. Q u an tu m num bers associated with
the upper level of a transition are indicated w ith a prim e, those corresponding to the lower
4
level w ith a double prim e. I f the variation in V
across the vibrational band is negligible,
then a B oltzm ann plot o f ln[I/(Sj<j-)] versus BV-J’(J’+ I) produces a line o f slope -hc/(kTr)
from which the rotational tem perature is determ ined. T he above assum es that the radiative
decay rates for various rotational levels are the sam e. If a given vibrational-rotational level
is m ixed with a different vibrational-rotational level then the radiative decay rate o f that
level m ay be altered and the intensity o f the line m ay be unusually strong or weak in com ­
parison with those o f others in the band.
Equation 4.1 can be represented in a sim plified form by
4.2
upperfcT
where Eupper
upper represents the energy level o f the upper electronic state.
35
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
U nder typical operating conditions, the plasm a w ith a neutral gas density o f around
10 1 7 cm ' 3 is far from local therm odynam ic equilibrium . H ow ever we can expect that, at
this pressure, the collision rate is high enough to equilibrate the rotational modes o f the
long lived electronic excited states o f H 2 w ith the heavy particles kinetic one [8 ]. In this
case, the plasm a heavy particle tem perature can be determ ined from the rotational bands
o f the radiative excited states o f H 2. T he tem perature was determ ined using the R branch
o f the G l Z +— ► B 1Z
5
+ (0,0) electronic transition, w here G * Z + and B *Z
E
“
M
^denote
the upper and low er electronic levels involved in the transition. The (0,0) sym bol repre­
sents a band o f transitions occurring betw een the levels with
ber in the upper electronic state and the levels w ith
0
0
vibrational quantum num­
vibrational quantum num ber in the
low er electronic state.
Singlets
nscr
npa
ndcr
doubly excited
levels
120,000 — 4
100,000
_
2
-
50,000
Figure 4.1 D iagram of the O bserved Electronic States o f the H 2 M olecule
36
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The electron configuration for this transition is ls<r3da. T h e observed electronic
states o f the
m olecule is shown in Figure 4.1. A ccording to the selection rule for elec­
tronic transition, if both the upper and low er electronic state involved in the transition have
zero electronic angular m om enta, only the transitions with AJ = ± 1 are allowed, w here AJ
is the change in the rotational quantum number. The AJ = +1 transition gives rise to the R
branch and AJ = -1 transition gives rise to the P branch. Since the electronic angular
m om entum o f a E s ta te is zero, th e G 1 2
g
+— ► B 1Z Li + (0,0) electronic transition has the
R and P branches as shown in Figure 4.2
r
8
7
6
G 1^
5
3
0
r
8
7
ErZ u
6
5
3
0
R
P
J ’ ' 7 4 2 O' '1 2 3 4 5 6
7
8^
Figure 4 .2 Energy Level D iagram for a Band with P and R B ranches
37
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The B 1 Z w + level is well described by H und’s lim iting case b approxim ation [A
brief discussion about the H und’s coupling cases is included in Section 4.2.1]. This is not
the case for the G 1
level w hich is interm ediate betw een the H und’s lim iting cases b
and d [9,10]. T hen, the rotational energy is no longer a linear function o f J(J+I). T here­
fore, the exact num erical values o f the rotational energy levels in the tem perature B oltz­
mann plot has to be considered [11,12]. A lthough these levels are perturbed by the high
vibrational levels o f the EFl
} +, the strength o f m ost o f the R branch rotational lines
can be w ell described by the HOnl-London form ulae, i.e. S^=(J+1) / 2.
Only ten em ission lines (R 0 --R io) were identified. T he R 1? R 4 and R2, R 3 lines are
not resolved and w ere not used in the B oltzm ann plot.
4.2.1. Hund’s Coupling Cases
The influence o f rotational and electronic m otions on each other is given by the
H und’s coupling cases a to e. The different angular m om enta in the m olecule - electron
spin S, electronic orbital angular momentum L o r A, angular momentum o f nuclear
rotation N - form a resultant that is designated J [13]. If S and A are zero, as in the
state, the angular m om entum o f nuclear rotation is identical w ith the total angular m om en­
tu m /.
In H und’s case a it is assum ed that the interaction o f the nuclear rotation with the
electronic m otion (spin as well as orbital) is very w eak, w hereas the electronic motion
itself is coupled strongly to the line jo in in g the nuclei. T he total electronic angular
momentum about the intemuclear axis Q. and the angular momentum N o f nuclear
rotation form the resultant / .
38
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Case b describes the condition when A = 0 and S £ 0 , the spin vector 5 is not
coupled to the intem uclear axis at all. Som etim es, particularly for light m olecules, even if
A ^ 0 , S may be only very w eakly coupled to the intem uclear axis. In this case the angu­
lar m om enta A and N form a designated K, w here K is the total angular momentum apart
from spin.
Case c discusses certain cases like heavy m olecules, w here the interaction between
L and S may be stronger than the interaction w ith the intem uclear axis. In this case L and S
first form a resultant Ja w hich is then coupled to the intem uclear axis w ith a com ponent Q.
Q and N then form the resultant angular momentum J.
Case d arises if the coupling between L and the intem uclear axis is very weak
while that between L and the axis o f rotation is strong. In this case the angular m om entum
o f nuclear rotation w hich is called R (rather than N) is quantized. T he angular momenta R
and L are added vectorially, giving the total angular momentum apart from spin, desig­
nated by K.
Case e occurs w hen L and S are strongly coupled. L and 5 form a resultant Ja which
is com bined with R to form J.
H und’s coupling cases represent idealized limiting cases. N evertheless they do
often represent the observed spectra to a good approximation. H ow ever, sm all o r even
large deviations from these lim iting cases are observed. These deviations have their origin
in the fact that interactions w hich w ere neglected or regarded as sm all in the idealized cou­
pling cases really have an appreciable m agnitude, and particularly that the relative m agni­
tude of the interactions changes w ith the increasing rotation. T herefore, som etim es, with
increasing rotation, a transition takes place from one coupling case to another.
39
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4.3 Experimental setup
T he m easurem ents w ere perform ed in the high pow er m icrow ave cavity plasm a
reactor shown in Figure 2.4. The collection optics arrangem ent #3 was used to focus the
optical em ission from the discharge to the spectrom eter system #! described in chapter 2 .
A spectrom eter w ith a resolution o f 0.2 A for a slit width o f 10 [I m was used for the m ea­
surements. The entrance and exit slits o f the spectrom eter were set to 50 p. m wide by 2cm
high in order to get the optim um spectral resolution and signal intensity.
Microwave Input Probe
Monochromator
Photomultiplier
Sliding Short
f=15cm
Window
f=30cm
f=5cm
Plasma
IEEE-488 bus
Quartz Dome
Substrate Holder
Air Coolin
Computer
/ Picoammeter
Figure 4.3 E xperim ental Set-up for the m easurem ent o f rotational tem perature o f
H2
40
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
T he optical em ission spectroscopy experim ents were carried out using a M cP h er­
son 0.5 m, plane grating scanning m onochrom ator with a 2400 grooves/m m grating an d a
EG I-G EN C O M RPI Q L/20 photom ultiplier tube. A voltage o f -800 V was applied to the
photom ultiplier tube. The output o f the photom ultiplier tube was connected to the K eithley 485 A utoranging picoam m eter, w hich w as interfaced to a com puter using the EEEE488 interface. The data acquisition and processing w ere perform ed by the computer. T he
com plete source code o f the Q B A SIC program used in the processing is included in
A ppendix C.
4.4 Hydrogen rotational temperature
A series o f experim ents w ere perform ed using pure H 2, a m ixture o f H 2 and C H 4
and a m ixture o f H 2, CH 4 and N 2. The param eters that were varied include input pow er,
pressure and flow rate o f the gases. The m onochrom ator was scanned in the range o f 453 0
A to 4650 A, corresponding to the R0-R10 rotational band o f H 2 m olecule.
4.4.1. Estimation of the rotational temperature
T he calculation involved in the determ ination o f the rotational tem perature is o u t­
lined below for the H 2 -C H 4 plasm a operated at 30 Torr pressure, 200 seem flow rate [H 2],
4 seem flow rate [CH4] and a m icrow ave input pow er o f 400 W. The microwave input
power m eans the resultant pow er input to the system , i.e. the difference between the in c i­
dent pow er and the reflected power. Figure 4 .4 shows the em ission spectrum obtained fo r
the above said operating conditions. The various rotational lines are m arked on the plot.
As it could be seen, the R l, R 4 and R2, R3 lines are not resolved. H ence they are not u sed
in the calculations.
41
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
R l, R4
e
3
R 2, R3
[R8
C /3
c
a>
c
R6
R5
RIO
R9
4540
R7
4580
4560
RO
4620
4600
4640
Wavelength A
Figure 4.4 Em ission Spectra o f the R -B ranch rotational lines o f H 2
R5
-3.5
RO
R7
R6
R8
R9
-4.5
-5.5
RIO
500
1000
1500
2000
2500
Relative upper level energy (cm )
Figure 4.5 Boltzm ann plot fo r the lines R0 and R5-R 10
42
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3000
Table 4.1: Energy level for the R-branch rotational lines
Rotational
Line
W avelength
o
A
Relative
upper level
energy cm '
S
1
R0
4627.5
292.86
60.01
R5
4624.7
895.24
86.17
R6
4618.4
1150
315.83
R7
4598.1
1490.48
113.09
R8
4581.3
1835.71
356.69
R9
4557.4
2238.1
199.34
R10
4537.9
2666.67
642.39
Table 4.1 shows the upper energy level for the R -branch rotational lines and the
corresponding value o f S, the H O nl-London factor. The value o f ln(I/S) is calculated and
plotted against the upper level energy, as shown in Figure 4.5. T h e line o f best fit is
he
obtained for the plot. The slope o f this line corresponds to —-— , w here c is the speed o f
rC
L
light in cm/s. From the value o f the slope, the rotational tem perature o f H 2 is obtained.
4.5 Rotational temperature results
The experimental results for the rotational tem perature are presented below. The
accuracy of rotational tem perature determ ined using this m ethod is found to be within
±200 K. This is estim ated from the reproducibility o f the data obtained.
From Figure 4.6 it is seen that the rotational tem perature Tr o f H 2 increases with
pressure. The tem perature ranges from 1200-2000 K. Also from Figure 4.7 it can be
observed that Tr of H 2 in H 2 -C H 4 m ixture is higher than that o f a pure H 2 plasm a dis-
43
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
charge. Figure 4.8 shows the increase in Tr with increase o f input power. As more pow er is
fed to the system , i.e. more energy is used to h eat the gas and hence the kinetic energy o f
the gas m olecules increases resulting in the increase o f Tr
W hen the flow rate o f the gas through the system was varied w ith the other param ­
eters held constant, a slight decrease in Tr was observed w ith increasing flow rate, espe­
cially at the low er rates below 100 seem . W ith higher flow rate the am ount o f time that a
gas m olecule spends in the active region o f the plasm a decreases as m ore molecules flow
through the system . Consequently, the m olecules do not gain as m uch kinetic energy as
com pared to the case when the flow rate is less, w herein the gas m olecules remain in the
active region o f the plasmas for a longer time. H ence the rotational tem perature Tr
decreases slightly as shown in Figure 4.9.
Since the presence of nitrogen plays a m ajor role in the diam ond deposition pro­
cess [14], it is interesting to determ ine its effect on the rotational tem perature o f the gas.
N itrogen was introduced into the gas m ixture an d its flow rate was varied a small am ount
as com pared to the total flow rate present in the actual diam ond deposition system. As
seen in Figure 4.10, there is no significant change in Tn indicating the absence o f influence
o f 0 . 1 - 1 % nitrogen in determ ining the rotational tem perature o f the given gas mixture.
44
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2400
Rotational Temperature Tr (K )
2200
2000
1800
1600
o
1400
1200
1000
0
10
20
40
30
Pressure (Torr)
50
60
F igure 4.6 Variation o f rotational tem perature o f H 2 w ith pressure for H 2 plasm a.
Flow rate: H 2-200 seem . Input power: 400 W
45
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2400
Rotational Temperature Tr (K )
2200
2000
1800-
1600-
1400-
1200
-
1000
30
40
Pressure (Torr)
Figure 4.7 Variation o f rotational tem perature of H 2 w ith pressure for H 2 -C H 4
plasm a. Flow rate: H 2-200 seem , C H 4-4 seem . Input power: 400 W
46
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2400
Rotational Temperature Tr (K )
2200
2000
-
-
1800
1600
1400-
1200
1000
200
300
400
Input Power (W)
500
600
Figure 4.8 Variation o f rotational tem perature o f H 2 w ith input pow er for H 2 -CH 4
plasma. Flow rate: H2-200 seem , CH4-4 seem. Pressure: 30 Ton-
47
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2400
Rotational Temperature Tr (K )
2200
2000
1800-
1600
1400-
1200
1000
-
100
150
200
250
300
350
400 450
Flow rate (seem)
Figure 4.9 V ariation o f rotational tem perature o f H 2 with flow rate for H 2 plasm a.
Pressure: 30 Torr. Input pow er: 400 W
48
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2400
Rotational Temperature Tr (K )
2200
2000
-
1800-
1600-
1400-
1200
-
1000
Flow rate (seem)
Figure 4.10 Variation o f rotational tem perature o f H 2 w ith flow rate of N 2 fo r H 2CH 4 -N 2 plasm a. Flow rate: H 2-200 seem , C H 4 -4sccm . Pressure: 30 Torr. Input pow er: 400
W
49
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4.6 Conclusion
The rotational tem perature o f H 2, H 2 -C H 4 and H 2 -CH 4 -N 2 microwave plasm a dis­
charge w ere measured. The tem perature was m easured to an accuracy o f +200 K. The
results show ed that both increase in pressure and increase in m icrow ave input power
increase the tem perature.
The results also show ed that the rotational tem perature varies only slightly with
the gas flow rate. The variation in the rotational tem perature w ith nitrogen concentration
w as o f particular interest, since small am ount o f nitrogen [0-25 ppm] present in the gas
m ixture has been found to enhance the grow th rate in CVD diam ond deposition [15]. But
w ith the
0 .1
- 1 % addition o f nitrogen no change in the rotational tem perature was
observed. This concludes that either the change in rotational tem perature due to N 2 con­
centration was w ithin the error limit, o r there was no change at all.
50
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Chapter 5
Study of H 2 - CH4 - N 2 Microwave Plasma
5.1 Introduction
This chapter m easures using OES the influence o f nitrogen on H 2 -CH 4 diam ond
deposition plasm as. The changes in various species em issions from the plasm a are m ea­
sured as nitrogen o f controlled am ounts is added. The m otivation for this set o f m easure­
ments is that in the plasm a-assisted chem ical vapor deposition o f diam ond films, the
addition o f sm all am ounts o f nitrogen changes the deposition rate and the characteristics
of the diam ond film [1, 14, 15]. The reason for this strong influence is as yet not under­
stood. One possible influence o f the N 2 addition is it produces changes in the bulk plasm a
properties and another reason is the nitrogen has a deposition surface influence. The m ea­
surem ents taken here by OES is directed at looking for bulk plasm a species changes. In
the previous chapter the influence o f N 2 addition to diam ond deposition plasm a gas tem ­
perature was m easured. The result was that no detectable gas tem perature change was
observed as nitrogen was added in small amounts (less than 2% ). The first section o f this
chapter gives a brief detail of the experim ental setup. The second section presents the
results and a discussion o f the results. The results are sum m arized in the concluding sec­
tion.
5.2 Experimental Setup
The experim ents were perform ed in the high pow er m icrow ave cavity reactor as
shown in Figure 2.4. The 0.5 m eter spectrom eter with 1200 lines/m m grating was used.
The optical em ission from the discharge was collected using a lens arrangem ent as shown
51
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
in Figure 4.3. T hree lenses w ere used to focus the optical em ission from the discharge on
to the entrance slit o f the spectrom eter. The lens arrangem ent is show n in detail in Figure
4.4. T he entrance and exit slits o f the spectrom eter w ere set to 50 (4 m w ide by 2 cm high.
The setup is identical to that used in chapter 4.
F or the Ha and the Hp m easurem ents, the spectrom eter was not scanned over the
w avelength range, as was done in other experim ents. Instead, the spectrom eter was set to
the w avelength corresponding to the peak of the H a line at 656.28 nm (486.13 nm for Hp).
W ith the spectrom eter fixed at this position the flow rate o f N 2 was varied to give N 2 c o n ­
centration o f 0-5% . The corresponding variation in the intensity o f the signal from the
picoam m eter w as recorded by the com puter and plotted. The C H and C N intensity m ea­
surem ents w ere obtained by scanning over a range o f 3800 to 4400 A°. The CH band at
4300 A 0 and the C N band at 3880 A 0 were used for the m easurem ents are shown in Figure
5.4.
5.3 Results and discussions
From Figures 5.1 and 5.2 it is observed that the intensity o f the atom ic hydrogen
(H a and Hp) increases w ith the percentage of N 2 in the gas m ixture. T he increase in the H a
and Hp em ission is by a factor o f greater than 3 as the am ount o f N 2 is increased from 0 to
5%. T he increase in the H a and Hp emissions could occur due to eith er an increase in the
atom ic hydrogen concentration [H] o r an increase in the plasm a electron tem perature. A
higher electron tem perature in the plasm a w ould m ean that a larger portion o f the elec­
trons are capable o f exciting the atom ic hydrogen to the H (n=3) and H (n=4). These are the
states that the H a and Hp em issions originate from.
52
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The ratio o f H p/Ha for different percentage o f N 2 as shown in Figure 5.3 gives an
idea of the change in electron tem perature with N 2 flow. The Hp line involves the transi­
tion o f electrons from the n=4 to n=2 state and the H a line involves the transition o f the
electrons from the n=3 to n=2 state. H ence an intense Hp line is indicative o f the presence
o f high tem perature electrons in the gas. The electron tem perature shows only a sm all
change as indicated by the ratio H p/Ha changing less than 4% . Hence, the strong changes
in Ha and Hp em issions is believed to be due to increases in atom ic hydrogen concentra­
tion as the N 2 is added, rather than a change o f the electron tem perature.
A typical C N and CH em ission spectrum is show n in Figure 5.4. The C H line
intensity shows a dependence sim ilar to that exhibited by the hydrogen lines versus N 2
concentration as show n in Figure 5.5. However the dependence is not linear o r as large as
shown by the hydrogen lines. The C H em ission intensity increases by a factor o f approxi­
m ately 2 as the N 2 concentration changes from 0-5% . T he C N line intensity shows a larger
variation with N 2 com pared to the o th er species. T he C N radicals as shown in Figure 5.6
tend to be prom oted by adding nitrogen to the reactant gases. This could be attributed to
the reactions o f nitrogen with m ethane and carbon atom s in the gas phase, w hich form ed
the CN radicals. A lternatively, atom ic nitrogen m ay rem ove carbon atom s from the grow ­
ing surface and thus produce CN radicals [16]. T he plots o f C H and CN show that the
addition of N 2 to the plasm a in am ounts o f less than 1% produce changes in the C H and
C N concentrations in the plasm a.
53
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2.2
1.8
§
1.6
-D
oo
C
<u
a
0.8
0.6
0
1
2
3
4
5
% o f N2
Figure 5.1 Variation o f Ha line intensity with N 2 concentration. Flow rates: CH 4 4 seem, H 2-200 seem . Pressure: 30 Torr. Input Power: 0.8 kW
54
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1.6
X)
1.2
GO.
0.8
0.6
0.4
0
1
2
3
4
5
% of N2
Figure 5.2 Variation of Hp line intensity w ith N 2 concentration. Flow rates: C H 4-4
seem, H2-200 seem. Pressure: 30 Torr. Input Power: 0.8 kW
55
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
9
8.8
8.6
8.4
8.2
CO
8
7.4
7.2
7
0
1
2
3
4
5
% ofN2
Figure 5.3 Variation o f Hp / Ha ratio w ith N 2 concentration. Flow rates: CH4-4
seem , H 2-200 seem . Pressure: 30 Torr. Input Pow er: 0.8 kW
56
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Emission intensity (arb. units)
C N 3880 A'
C H 4300 A'
3800
3900
4000
4100
4200
Wavelength (A°)
4300
4400
F igure 5.4 Em ission spectrum o f H 2 -C H 4 -N 2 plasm a. Flow rates: CH 4 -7.2 seem ,
H o-144 seem , N 2 -0.5 seem. Pressure: 30 Torr. Input Power: 0.8 kW
57
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2
1.8
CH line intensity (arb. units)
1.6
1.4
1.2
1
0.8
0.6
0.4
0.2
0
0
2
% o f N?
3
Figure 5.5 Variation o f CH line intensity with N 2 concentration. Flow rates: CH 47.2 seem , H2-144 seem . Pressure: 30 Torr. Input Power: 0.8 kW
58
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
300
250
5 200
a 150
2 100
0
1
2
% o f No
3
4
5
Figure 5.6 Variation o f C N line intensity with N 2 concentration. Flow rates: C H 47.2 seem, H 2-144 seem. Pressure: 30 Torr. Input Power: 0.8 kW
59
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5.4 C o n clu sio n
The results presented above show a strong dependence o f Ha , Hp, CH and CN
em issions on the N 2 concentration. A n increase in the am ount o f atom ic hydrogen is
observed with increase o f nitrogen. Since it is believed that atom ic hydrogen plays an
im portant role both in the gas phase and on the grow ing surface during diam ond deposi­
tion, our observation shows that N 2 does change the bulk p lasm a properties. The Hp/Ha
ratio shows no significant change in the electron tem perature as the nitrogen is added to
H 9 -C H 4 plasm a. T he variation o f C H shows a sim ilar trend as th at o f the atomic hydrogen.
60
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Chapter 6
Study of Ar - H 2 - CH4 - N 2 Microwave Plasma
6.1 Introduction
This chapter extends the analysis presented in ch ap ter 5 to A r - H 2 - CH 4 - N 2 plas­
mas used to deposit diam ond. The addition o f noble gases such as A r has been found to
influence the grow th rate o f diam ond, and they were found to have a profound effect on
the plasm a chemistry, including ionization and dissociation [16]. This experim ent is aimed
at studying Ar - H 2 - C H 4 plasm as and the effect of nitrogen on A r - H 2 - CH 4 plasm as.
The motivation for studying this plasm a stem s from the fact that the surface morphology,
the grain size and the grow th m echanism o f the diam ond film can be controlled by varying
the A r/H 2 ratio in the A r - H 2 - CH 4 plasm a [2, 17]. The results o f chapter 5 show that the
addition o f nitrogen varies the am ount o f atomic hydrogen in the plasm a. Hence it would
be interesting to observe the effect in the A r - H 2 - CH 4 - N 2 plasm as.The first section
gives an overview o f the experim ental setup. The subsequent section describes the results
and presents a discussion o f them. The concluding section highlights the significant results
obtained.
6.2 Experimental setup
The experim ents w ere carried out in the high pow er m icrowave reactor shown in
Figure 2.5. This system is sem i-autom ated so that the flow o f the gases and the pressure
are m onitored and controlled by a computer. M icrowave pow er used in the experim ent is
in the range of 1-1.7 kW. This resulted in the formation o f a very bright plasm a, which
elim inated the need fo r any optical arrangem ent to focus the em ission into the core o f the
61
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
fiber. The optical arrangem ent #1 shown in Figure 2.8 was used. The signal was detected
using the spectrom eter system #2 shown in Figure 2.7. The input slit width was set to 0.4
mm. The exposure time o f the diode array detector was varied to prevent saturation o f the
detector. T he m axim um exposure tim e needed was 0.328 - 4 seconds. The center w ave­
length was varied by turning the grating o f the diode array detector manually.
6.3 Results and discussion
The em ission from the plasm a w as scanned at three regions, w hose center w ave­
lengths are 4 1 0 ,4 8 5 and 650 nm. The m ain em ission lines th at were observed are C H , CN,
C2, Ha and Hp. In the case o f CN and C 2, a band was observed due to the presence o f var­
ious rotational and vibrational levels associated with the m olecules. Since the exposure
time of the diode array detector was varied throughout the experim ent to prevent the satu­
ration o f the detectors, the results presented here are norm alized to an exposure tim e o f
1
second. For the C N and C 2 molecules, both the peak intensities and the area under the
band are presented after taking background noise into consideration. The species identi­
fied and their w avelengths are as follows: CN (383 nm), C H (430 nm), C 2(472 nm , 516
nm), Hp (486 nm ) and Ha (656 nm), sam e as those studied by Clay et al. [18]. The C 2
bands at 472 nm and 516 nm are labeled as C2_a and C2_b respectively. For area m easure­
ments, the suffix “area” is attached to the species nam e. 2% nitrogen gas was used in the
experim ents. The process values and the associated data file nam es are provided in A ppen­
dix B.
The source param eter that was analyzed included the gas flow rate o f H2, C H 4, N 2
and An The Figures 6.1, 6.2 and 6.3 show the em ission spectrum with center w avelengths
o f 410, 485 and 650 nm. Figures 6.4 and 6.4a show the variation o f the various species
62
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
concentrations with hydrogen flow. The concentration o f all the observed species tend to
decrease with increase in flow o f H 2. A lthough at high H 2 flow rates Hp and C H increase
slightly, CN radical concentration was observed to decrease by a factor o f 4 as H 2 flow
rate varies from 0-100 seem . This is follow ed by the C 2 w hich decreased by a factor o f 3
for the sam e flow rate range. The presence o f C N radical in the spectrum though nitrogen
was not added to the gas m ixture, can be attributed to the im purity present in the input
gases or in the cham ber. T h e result suggest that a lower am ount o f C H 4 dissociates to form
C 2 or CH with increased flow o f H2.
Figures 6.5 and 6 .5 a represent the variation o f Ha , Hp, C N and C 2 with m ethane
flow. Ha , Hp, CN and C 2 increases with the increase in flow o f C H 4, w ith C 2 and C H
increase being more prom inent. This result agrees with the general expectations o f an
increased carbon species w ith an increased flow o f methane.
Figure
6 .6
indicate the variation o f C N and CH with the concentration o f nitrogen.
C H remains unaffected by the changes in N 2 w hile CN shows a linear increase by a factor
o f 20 with increase in concentration o f N 2 by a factor o f 10. N itrogen form s C N readily
with increased flow, w hile n o t affecting the am ount o f CH in the system . This is in agree­
m ent with the results o f th e w ork done by Clay et al. [18] in a C H 4 /N 2 plasm a.T he C N
spectral lines are readily seen in the spectrum due to its low activation energy.
The influence o f argon on the concentration o f the other species is shown in Fig­
ures 6.7 and 6.7a. C 2 and H a increase by a factor o f 3.5 and 1.5 respectively w ith increase
in Ax concentration from 24- 92% . CN , Hp and C H show slight increase with the A r con­
centration.
63
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
In addition to the above plots, the ratios o f C 2 and Hp with H a are plotted in Fig­
ures 6 . 8 - 6 .11. The C 2 /H a ratio decreases with increase in H 2 flow and increases with
increase in CH 4 flow. The ratio o f Hp/Ha gives som e inform ation about the electron tem ­
perature o f the system. W ith increase in the flow o f H 2, the Hp/Ha ratio decreases. T he Hp/
Ha ratio increases steadily w ith the flow o f CH 4, suggesting a large increase in the elec­
tron tem perature with increase in m ethane concentration.
3500
3000
CN
388 nm band
2500
CH
430 nm band
2000
C/3
cu
<
c 1500
o
m
1000
500
370
380
390
400
410
420
430
440
W avelength (nm)
Figure 6.1 Em ission spectrum . Flow rates: A r- 600 seem, C H 4- 3 seem, H 212sccm, N 2- 3 seem. Pressure: 120 Torr. Input Pow er: 1.09 kW
64
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
450
400
Emission intensity
(arb.
units)
350
300
512, 516 nm band
250
200
472 nm band
150
Hr
486 n m
100
450
460
470
480
490
W avelength(nm )
510
500
Figure 6.2 Em ission spectrum . Flow rates: Ar- 100 seem, CH 4-
8
300sccm , N 2- 0 seem. Pressure: 120 Torr. Input Power: 1.621 kW
65
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
seem , H2-
900
656 nm
800
Emission intensity (arb. units)
700
600
500
400
300
200
100
620
630
640
650
660
670
680
W avelength (nm)
Figure 6.3 Em ission spectrum . Flow rates: Ax- 100 seem , C H 4-
8
300sccm, N 2- 0 seem. Pressure: 120 Torr. Input Power: 1.621 kW
66
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
seem , H2-
2-a area'
2-b
10 '
zA
100
40
120
140
160
180
2 0 0
H t flow rate (seem)
Figure 6.4 Variation o f species concentration w ith H 2 flow rate. Flow rates: Ar600 seem , C H 4-
8
seem . Pressure: 120 Torr. Input Power: - 1 . 2 kW
67
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
45
-
CN
Emission intensity (arb. units)
area
0
10
20
30
40
50
60
70
80
90
100
H2 flow rate (seem)
Figure 6.4a V ariation o f species concentration with H 2 flow rate. Flow rates: Ax600 seem, CH4-
8
seem . Pressure: 120 Torr. Input Power: - 1 . 2 kW
68
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
-
CN
area
14-
CH
5 12:
2-a area
•2 -b
CO
cJ
C
c
c
o
*co
|
a
6
^
2
k?-
20
24
CH 4 flow rate (seem)
Figure 6.5 Variation o f species concentration with CH 4 flow rate. Flow rates: Ar600 seem, H2- 0 seem . Pressure: 120 Torr. Input Power: ~ 1.2 kW
69
Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
0 . 6'
0.3
CN
C
^ N area
CH
c
3
c
o
c
i 0.2
f=
a
0.H
C H 4 flow rate (seem )
Figure 6.5a Variation o f species concentration w ith C H 4 flow rate. Flow rates: Ar600 seem, H2- 0 seem . Pressure: 120 Torr. Input Pow er: - 1 . 2 kW
70
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2.5'
Emission Intensity (arb. units)
CN
CN
'“
i>area
CH
1.5
0.5
N 2 flow rate (seem)
Figure
6 .6
Variation o f species concentration with N 2 flow. Flow rates: A r- 600
seem , H 2- 12 seem, CH4-
8
seem . Pressure: 120 Torr. Input Power: ~ 1.1 kW
71
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1.4
1.2
2-a
2-a area
e
3
2
.p
-b
i°.8
*c/5
c
0
1 0.6
.2
*55
c/5
£
w
0.4
0.2
50
70
80
90
100
% Argon
Figure 6.7 Variation o f species concentration with A r flow rate. Flow rates: H 2 50-300 seem , CH4-
8
seem . Pressure: 120 Torr. Input Power: - 1.4 kW
72
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0 .0 3
CN
CN area
CH
0.025
0.02
0.01
0.005
90
40
100
% Argon
Figure 6.7a Variation o f species concentration with A r flow rate. Flow rates: H 2 50-300 seem, CH4-
8
seem . Pressure: 120 Torn Input Power: ~ 1 .4 kW
73
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
C2 -a ^ a
^-■2-a a r e a t a
Intensity ratio
^ 2-t/H a
40
Figure
C H 4-
8
6 .8
60
50
100
120
H 2 flow rate (seem)
140
160
180
2 0 0
Variation o f C 2 /H a ratio with H 2 flow rate. Flow rates: Ax - 600 seem ,
seem . Pressure: 120 Torr. Input Power: ~ 1.2 kW
74
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Intensity ratio
40
CH 4 flow rate (seem )
Figure 6.9 Variation o f C 2 /H a ratio with CH 4 flow rate. Flow rates: A r - 600 seem,
Ho- 0 seem . Pressure: 120 Torr. Input Power: ~ 1.2 kW
75
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0 .0 8
0.07
Intensity ratio
0.06
0.05
0.04
0.03
0.02
0.01
40
10
100
120
140
160
180
200
Ho flow rate (seem)
Figure 6.10 Variation o f Hp/Ha ratio w ith Ho flow rate. Flow rates: A r - 600 seem,
CH4-
8
seem . Pressure: 120 Torr. Input Power: ~ 1.2 kW
76
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0 .1 4
H r/H (
0.12
0.1
0.06
0.04
0.02
22
24
C H 4 flow rate (seem)
Figure 6.11 Variation o f Hp/Ha ratio w ith PI2 flow rate. Flow rates: A r - 600 seem ,
H 2- 0 seem . Pressure: 120 Torr. Input Pow er: - 1 . 2 kW
77
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6.5 Conclusion
T he em issions from several species including H, CH, C N and C2 were m easured
versus variation in source operating param eter including input gas flow com position. The
results indicate that in AX-H2 -C H 4 plasm as the increase in argon flow percentage produces
m ore C 2 species in the plasm a. T he C 2 species is im portant since it is believed to be a key
precursor for nanocrystalline grow th [2]. The am ount o f argon was found to affect the
atom ic hydrogen em ission.
T he influence o f nitrogen on the A r-H 2 -C H 4 plasm as was also examined. The
am ount o f N 2 added was 0-300 ppm . As expected the CN em ission was seen to depend
strongly on the N 2 concentration. The C H em ission was not affected by nitrogen concen­
tration.
78
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Chapter 7
Summary of results
7.1 Conclusions
The experim ents were perform ed in accordance w ith the objectives, which is to
investigate the m icrow ave plasm a discharges using optical em ission spectroscopy. The
w ork was started w ith the aim o f studying the A r p lasm a and m easuring the amount o f
doubly ionized atom s, and studying the variation o f gas tem perature o f H 2 - CH 4 micro­
w ave plasm a and the effect of N 2 addition to the plasm a. T he above objectives were real­
ized by perform ing various experiments as discussed in the preceding chapters. This
ch ap ter presents a b rie f summary o f all the results obtained and how they have helped in
the realization of the goals set forth in chapter 1 .
7.1.1 Argon Doubly-ionized atoms measurements
The experim ents were carried out in a com pact ion source. The results confirmed
the presence of Ar*-1" ions in the com pact ion source. Experim ents to determ ine the depen­
dence o f the Ar++ em ission on the source operating conditions were perform ed. The Ar++
ion density was found to increase as the flow rate and the pressure decreases. The results
presented gives an idea o f the A r ^ density variation in a com pact ion source. And, it pro­
vides a relative indication of the high energy (>28eV ) electron density.
79
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
7.1.2 Gas Kinetic Temperature of H2 - CH4 Microwave Plasma
The gas tem perature m easurem ents are im portant in understanding the plasm a
since the gas temperature helps determ ine the concentration o f various radicals. T h e gas
tem perature was obtained from the rotational tem perature, which was derived from the
m easurem ent o f the relative intensities o f rotational lines w ithin a single vibrational band.
T he tem perature obtained ranges from 1200-2000 K. The rotational tem perature is found
to vary slightly with the gas flow rate. T he rotational tem perature tends to increase with
the pressure o f the gas m ixture. Interestingly, the 0.1 - 1% addition o f nitrogen d id not
change the rotational tem perature by an observable value.
7.1.3 Study of H2 - CH4 - N2 Microwave Plasma
T he experiments w ere perform ed in conditions sim ilar to that o f an actual diam ond
deposition system . Typical values o f input power, pressure and flow rates w ere 0.8 kW, 30
Torr and 200 seem, 4 seem and 0-5 seem for H2, C H 4 and N 2 respectively. B asically the
variation o f H a , Hp, CH and C N em issions produced by N 2 concentration variations was
observed.
The results show a strong dependence o f the observed species on the N 2 concen­
tration. An increase in the am ount o f atom ic hydrogen is observed with increase o f nitro­
gen. The Hp/Ha ratio shows alm ost no change in the electron tem perature o f the m ixture
with the inclusion of N2. T he variation o f CH shows a sim ilar trend as that o f the atom ic
hydrogen.
80
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
7.1.4 Study of Ar-H2 - CH4 - N2 Microwave Plasma
The Ar-H-> - C H 4 - No m icrowave plasm a is used in the deposition o f nanocrystalline and m icrocrystalline diam ond thin films. T he experim ents were carried out in the high
pow er m icrow ave reactor. The typical values o f pow er, pressure and the flow rates o f H 2,
Ar, CH 4 and 2% N 2 in H 2 are 1.8 kW, 120 Torr, 0-300 seem , 600 seem , 3-24 seem and 010 seem respectively. Som e significant results that w ere obtained are m entioned here.
T he increase in the argon flow produces m ore C 2 in the plasm a. The C 2 species is
im portant since it is believed to be a key precursor fo r nanocrystalline grow th. T he am ount
o f argon was also found to affect the atom ic hydrogen em ission. U nlike the previous
experim ents the CH em ission was not affected by the nitrogen concentration change in the
0-300 ppm range.
7.2 Recommendations for future work
This study has contributed to the understanding o f a few plasm as used in diam ond
deposition and low pressure plasm a processing o f m aterials. The results obtained w ere
within o u r expectations. B ut som e anom alies are found to exist, which need further careful
study. O ne such exam ple is the variation o f A r4-1- density with power. Even though, the
Ar4-1- density was supposed to increase with pow er, it tends to saturate at higher po w er as
shown by the experim ents. A nother study o f interest w ould be the com parison o f the tem ­
perature m easurem ents w ith other m ethods like doppler broadening. A n attem pt w as made
to m easure the sam e using a Fabry-Perot cavity, but due to inadequate signal, the m easure-
81
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
m ents w ere not carried out. For a more com plete understanding o f the plasm a species co n ­
centration, the plasm a can be m odeled theoretically and the plasm a param eter variations
predicted by the m odel can be used to com pare th e validity o f the experim ental results. F o r
exam ple, the plasm a param eters in the low pressure source are found to exhibit a large
dependence on the flow rate o f the process gases. A theoretical model should be able to
help explain this dependence. Also to understand the variation o f species concentration
and the plasm a param eters w ith time, the OES experim ents can be perform ed in an actual
diam ond deposition process. This w ould also facilitate the study o f the characteristics o f
the deposited diam ond film under the various processing conditions.
82
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
APPENDICES
83
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
APPENDIX A
Table : A Identification and analysis of argon doubly ionized atoms
Input
Pow er
(W )
Pressure
(mT)
Flow rate
(seem)
1
40
0.3
2
40
3
Expt. #
In ten sities (arbitrary units)
3336.13A
3344.72A
3358.49A
1
7 .2 9 e -ll
5 .9 4 e-l 1
6 .0 5 e-l 1
0.5
1
6 .5 5 e -U
6 .3 8 e-l 1
5 .10e-l 1
40
0.75
1
4 .5 0 e -l 1
3 .0 9 e -ll
3 .1 le - 1 1
4
40
0.9
1
4 .9 0 e -l 1
3 .0 9 e -Il
2
5
40
0.9
8
3 .1 6 e -ll
1 .7 4 e -ll
2 .4 8 e -ll
6
40
0.9
2 0
0
0
5.68e-12
7
40
2
1
2 .9 0 e-l 1
1 .4 6 e -ll
8
40
3
1
1.04e-l 1
5 .1 4 e -ll
9 .1 2 e -l1
9
40
3
8
1 .1 5 e -l1
5.99e-12
1.24e-l 1
1 0
40
3
2 0
0
0
7.20e-12
2 1
1 0 0
0.3
1
8 .8 1 e -ll
5 .7 4 e -ll
6 .9 3 e -l1
2 2
1 0 0
0.5
1
5 .7 7 e -ll
3 .6 9 e -ll
4 .3 le - 1 1
23
1 0 0
0.75
1
4 .1 8 e - ll
4 .5 9 e -ll
4 .3 9 e-l 1
24
1 0 0
0.9
1
4 .9 6 e - ll
3 .4 2 e -ll
4 .2 2 e -ll
25
1 0 0
0.9
8
2 .9 1 e -ll
2 .2 8 e -ll
2
26
1 0 0
0.9
2 0
5 .0 3 e -U
3 .4 3 e -ll
3 .8 8 e -ll
27
1 0 0
2
1
1 .5 9 e -ll
6 .2 3 e -ll
2 .04e-l 1
28
1 0 0
3
1
7 .6 7 e - ll
3 .1 5 e -ll
1.26e-l
29
1 0 0
3
8
4 .2 3 e - ll
3 .3 2 e -ll
8 .29e-l 1
30
1 0 0
3
2 0
1.28e-l
1.4 6 e-11
31
1 0
0.9
1
1 .5 6 e -ll
32
2 0
0.9
1
2
1
.6 6 e - l l
.8 6 e - l l
.
1 8 8
e -l
1
.6 8 e - l l
1
0
.1 2 e - l l
4 .8 8 e-l 1
3 .7 4 e -ll
l.4 8 e - ll
1
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table : A (contd.)
Input
Pow er
(W)
Pressure
(m T)
Flow rate
(seem)
33
30
0.9
34
50
35
Expt. #
Intensities (arbitrary units)
3336.13A
3344.72A
33 58.49A
1
3 .2 3 e -ll
2 .8 9 e -ll
1.72e-l 1
0.9
1
5 .8 9 e -ll
3 .0 9 e -ll
3 .4 9 e -ll
60
0.9
1
5 .7 le - 1 1
3 .6 8 e - ll
3 .4 2 e -ll
36
70
0.9
1
6 .1 5 e -ll
3 .3 5 e - ll
4 .4 2 e -l 1
37
80
0.9
1
5 .8 7 e-l 1
4 .4 6 e - ll
3 .4 4 e -l 1
38
90
0.9
1
6 .9 7 e-l 1
4.4 l e - 1 1
4 .5 9 e -l 1
Input Pow er = Incident Pow er - Reflected Pow er
Intensities w ere calculated by m easuring the area under the peaks at 3336.13, 3344.72 and
3358.49A.
85
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
S
u
3
cn
oa
jy
z
£
400
400
1075
1075
355
355
414
414
414
1897
1897
1878
1878
1878
OO
OO
OO
OO
86
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 0 0
OO
009
oT
o
o
600
o
1 2 0
1 2 0
1 2 0
1 2 0
1 2 0
1 2 0
1 2 0
1 2 0
1 2 0
1 2 0
1 2 0
1 2 0
Pressure
(Torr)
i
650
485
410
410
485
650
650
485
410
410
485
Exposure
time (s)
Filename
F4
F3
F2
F7
F6
F10
F9
F12
X
1038
o
1 0 0
o
1038
2 0 0
OO
2 0 0
OO
2 0 0
CO
o
IX
1038
400
1075
355
1897
GO
2 0 0
2 0 0
I
OQ
2 0 0
1 1 0 0
315
1936
2 0 0
o
OO
009
1 1 0 0
315
2 0 0
o
1936
300
o
2 0 0
cn
1936
300
o
1 1 0 0
1 0 0
1115
§
315
c
1936
y;
E1 0 0
u
O
1 1 2 0
o
300
E
315
X
C
v—'J
j?
1936
(M) l,3Jd
aV
Nf* O
C
->
c/2
o
o
O
ex
0 0 1
U
l
c*
X
t
u
<
oE
uo
1 0 0
01
1 1 2 0
Z
^-
315
CU
>
a
£
osu
u
J
1936
S3
Ar
(seem)
APPENDIX B
^
tx
■Tt*
o
C3
r*
c
V3
u
co
OO
OO
zI
F25
-'3-
Tf
F27
F24
CN
F26
F23
nj-
OO
CN
CL,
0.656
F22
F21
F20
F19
F17
F16
F15
CL,
■'3-
-'3-
in
00
0
in
vo
O
in
VO
in
OO
0
in
vO
O
in
vO
in
00
0
0
' 3-
■'d'
0
cn
0
01
0
01
0
01
0
CN
0
01
0
CN
0
CN
O
CN
O
CN
O
0
O
O
O
O
0
O
c
O
O
in
CN
in
Ol
CN
Ol
Ol
O
VO
VO
O
0
O
O
CO
00
OO
00
OO
OO
OO
OO
OO
00
00
OO
0
0
VO
0
0
O
O
O
O
0
0
0
VO
0
0
vo
0
0
vO
0
0
VO
O
O
VO
O
O
VO
0
0
vo
O
O
vO
O
O
VO
in
in
Cn
in
C
0
in
SO
O
VO
0
nt-
0
O
Cv
Cv
O
O
in
Cv
in
CN
O
Cv
00
OO
cn
cn
CN
cn
in
CN
cn
in
(M
cn
m
Cv
Cv
Cv
Cv
in
Cv
Cv
Cv
00
OO
1878
00
00
OO
00
1878
00
00
1878
"3-
00
00
in
00
0
in
vO
0
m
vO
in
00
O
0
01
0
cn
O
CN
O
c-i
0
01
0
01
0
0
O
O
O
0
in
O
m
m
CN
00
00
(se e m )
■rt-
1878
Exposure
time (s)
1
F14
F13
Filename
00
O
O
VO
O
O
0
O
O
vO
Pressure
(Torr)
<<g
o
0
(llIODS)
s'
o
00
X
u
Ar
%
o '
Vi
cn
r-
Cn
r^r
cn
r-*
Cv
Cv
'O
ON
O
CN
cn
Cv
C n
Cn
00
OO
00
00
00
00
^r
r~~
Cv
Cv
Cv
m
in
in
Cv
in
CV
in
C
in
CL
g
u
00
00
00
00
r-
r-
1878
Table: B (contd.)
§
1 0 2 0
ZH
0
in
1 0 2 0
£
1 0 2 0
<n
Z
r—
r»
O h"
87
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
r-
r-~
r-~
00
00
650
650
730
730
730
493
OO
1878
1878
OO
rOO
1878
OO
rOO
o
CN
OJ
Ol
Ol
Ol
Ol
24
"3-
600
600
2 0 0
2 0 0
2 0 0
CN
OO
OO
OO
CO
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
IT.
£ \
069
88
069
ClT
o
915
o
1 2 0
1 2 0
1 2 0
1 2 0
120
1 2 0
1 2 0
1 2 0
1 2 0
1 2 0
1 2 0
1 2 0
1 2 0
1 2 0
1 2 0
1 2 0
!
F38
F39
0.164
650
410
0.328
F44
F43
485
F42
F41
F40
410
485
0.656
F36
0.656
410
F37
F35
485
0.164
F33
F32
F34
i
0.328
F30
F31
F29
0.656
0.65
650
650
485
410
410
485
Filename
Exposure
time (s)
o
"3"
1878
o
o
1878
o
493
o
1 1 2 0
o
1 1 0 0
o
o
1878
uo
C\
OO
895
o
24
1
O
24
o
493
OO
895
1 0 1 0
1 0 1 0
1 0 1 0
o
1878
[
650
1878
o
o
O
Ol
OO
009
895
o
vC
009
069
690
o
om,
009
o
OO
rOO
m,
OO
009
ex
OO
P"*
OO
Ol
009
1
o
009
rVO
OO
Ol
009
O
o
009
§
o
009
y
009
1
o
895
o
o
009
"p
690
-r
1878
rS-"j/
895
X
o
009
u
1878
<M O
E
863
&
591
Z
591
E
1878
Pressure
(Torr)
^
009
^
(lUOOS)
Ar
(seem)
r -1
009
CN
ZE
Table: B (contd.)
rr
F-
965
oo
oo
r—
oo
00
1878
910
870
o
r—
730
OO
CO
V
ooO
836
820
001
oo
750
750
788
OO
r-
1878
1878
50
24
oo
1 2 0
1 2 0
1 2 0
1 2 0
1 2 0
1 2 0
1 2 0
1 2 0
1 2 0
1 2 0
1 2 0
1 2 0
Pressure
(Torr)
1 2 0
1 0 0
OO
600
2 0 0
2 0 0
OO OO
o
o
—
Ol
o
Ol
Ol
Ol
oo
cn
CO
CO
F51
F50
F49
F48
0.328
un
oo
89
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
00
410
lU
F60
F59
OO
ir;
410
410
OO
0*1
cn
o
485
F54
F53
F55
0.328
0.656
0.656
1
F46
F45
Filename
0.328
485
485
485
485
0.656
0.656
650
0.656
650
0.656
0.656
650
0.656
0.656
Exposure
time (s)
650
650
(nm)
X
ri/~i
tu
1 2 0
1 2 0
0
01
1878
1878
870
692
C*^
oo
un
670
00
OO
oo
oo
1878
631
965
591
1 0 0
OO
o
009
915
600
Ol
o
009
650
50
OO OO
o
vO
m
£L*
009
OO oo
t— oo
oo r890
24
OO
o
009
Cv
8 8 6
OO
o
009
710
o
009
cC
729
m
oo
oo
1878
OO
oo
o
iri
tu
009
729
oo
o
009
1878
oo
069
OO
o
o
V,
vO
009
0)
oT
ov->
O
009
g
069
g
1878
m
Ol
009
p
o
009
su'
oOO
o
009
- t
M
-H
00
o
006
Ar
(seem)
T?
Z
CJ
O
009
Table: B (contd.)
CN OZ E
c°
Ol 3
r^
Tj"
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table: B (contd.)
ch
Pine ( W )
Prefl ( W )
Ts (°C)
Ar
(seem)
(seem)
h2
(seem)
1878
788
818
600
3
1878
788
815
600
1878
788
814
1878
788
1878
% N2
(seem)
Pressure
(Torr)
12
3
3
12
600
3
813
600
788
810
1878
808
1878
1878
X
(nm)
Exposure
time (s)
Filename
1 2 0
410
4
F61
4
1 2 0
410
2
F62
12
5
1 2 0
410
2
F63
3
12
6
1 2 0
410
2
F64
600
3
12
7
1 2 0
410
2
F65
812
600
3
12
8
1 2 0
410
1
F6 6
808
810
600
3
12
9
1 2 0
410
1
F67
808
809
600
3
12
10
1 2 0
410
1
F6 8
Pjnc - Incident Microwave
4
Power
Preil - Reflected Microwave
Power
Ts - Substrate Temperature (Observed using an optical pyrometer)
2
X - Center wavelength of the Spectrometer
APPENDIX C
QBASIC program to read data from the picoammeter connected to the monochro­
mator :
‘A uthor : Jayakum aran Sivagnanam e ; D ate: July 1998
‘M onochrom ator Spec: 2400 lines/m m grating: offset= 23.5 angstrom
‘M cpherson M onochrom ator m ust be scanned w ith increasing w avelength ONLY
D EC LA R E SUB FILEW RTU ()
D EC LA R E SUB FILEW RTD (=)
‘SINCLUDE: ‘qbdecl4.bas’
C O N ST
CO N ST
CO N ST
C O N ST
C O N ST
C O N ST
C O N ST
C O N ST
C O N ST
C O N ST
C O N ST
C O N ST
C O N ST
C O N ST
C O N ST
C O N ST
black = 0
blue = 1
green = 2
cyan = 3
red = 4
m agenta = 5
brown = 6
w hite = 7
grey = 8
lightblue = 9
lightgreen = 10
lightcyan = 1 1
lightred = 12
lightm agennta = 1 3
yellow = 14
brightwhite = 15
CO M M O N resol!, cyc%, count!, pico%, i%, stain!, tottim % , sec% , buff$, tem p$, prev
s tl$ = “w aveln=[“
st 2 $ = “ value=[“
EN$ =
path$ = “c:\jay\spec\data”
10C L S
20 SC R EEN 0
30 LOCATE 1, 3
40 PRIN T “E nter the last few digits [m m ddtrialno] o f the data file” ;
91
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
50
60
70
80
IN PU T filenmS
P R IN T “ Is the filenam e correct [y/n]” ;
IN PU T confirm s
EF confirm s = “ n” G O TO 10
C A L L EBDEV(0, 3, 0, 10, I, I, pico% )
O P E N pathS + “\W ” + filenmS + “ .m ” FO R O U TPU T AS # I ’store w avelength value
PR IN T # l , s t l $
O PE N pathS + “\D ” + filenmS + “.m” FOR O U TPU T AS # 2 ’sto re current value
PR IN T #2, st2$
LO C A TE 2, 3
PR IN T “Values o f w avelength stored in
pathS + “\W ” + filenmS + “ .m” ; “ file”
LO C A TE 3, 3
P R IN T “Values o f current stored in
pathS + “\D ” + filenmS + “ .m ” ; “ file”
buffS = SPACE$(40)
tem pS = SPACE$(40)
cycle! = 4
count! = 1
prev = 1E -11
C O L O R green, black
LO C A TE 4, 3
P R IN T “Program used for the spectrom eter with 2400 lines/m m g ratin g ”
C O L O R w hite, black
LO C A TE 5, 3
P R IN T “E n ter the initial scan w avelength [A]” ;
IN PU T stain!
C O L O R yellow, blue
LO C A TE 7, 3
P R IN T “Set the counter to” ; stain! + (47/2)
C O L O R w hite, black
LO C A TE 9, 3
PR IN T “E n ter the final scan w avelength [A]” ;
IN PU T endln!
LO CA TE 10, 3
P R IN T “E nd point as observed from the counter [A]” ; endln! + (47/2);
LO CA TE 11, 3
P R IN T “E n ter the speed o f scanning drive [A/min]” ;
IN PU T speed!
LO CA TE 13, 3
P R IN T “M axim um resolution that can be obtained [A]” ; speed! / (cycle! * 60);
LO C A TE 15, 3
P R IN T “E n ter the resolution required [A]-(defauIt” ; speed! / (cycle! * 60); “A )” ;
92
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
IN PU T resol!
IF resol! = 0 TH EN resol! = speed! / (cycle! * 60)
sec! = resol! * (cycle! * 60) / speed!
sec% = sec!
IF (sec% - sec!) > .000001 T H E N
C O LO R w hite, red
LO CATE 16, 3
PR IN T “E rror in calculation is m ore likely to occur” ;
EN D IF
C O L O R w hite, black
LO CA TE 17, 3
PR IN T “D ata acquisition interval [seconds]” ; sec% ;
tem p% = endln! - stain!
IF tem p% < 0
LO CA TE 19, 3
PR IN T “Please scan the spectrom eter with increasing value o f w avelength”
G O TO 200
E L SE
LO CA TE 19, 3
tottim % = (endln! - stain!) * 6 0 / speed!
PR IN T ‘T o tal time taken [seconds]” ; tottim % ;
endtim % = tottim % / sec%
LO CA TE 21, 3
PR IN T “N o. o f samples to be taken” ; endtim % * cycle!;
LO CA TE 22, 3
C O LO R w hite, red
PR IN T “Set Scan drive scanning sw itch to H” ;
LOCATE 23, 3
PR IN T “Start the Program and the Scan drive sim ultaneously” ;
C O LO R white, black
LO CATE 24, 3
PR IN T “H it any key to start” ;
W H IL E INKEYS = ““
W EN D
C O LO R green, black
i% =
1
T IM E R ON
O N TIM E R (sec% ) GOSUB disp
CLS
LO CA TE 1, 1
PR IN T “Program reads data from Picoam m eter in” ; sec% / cycle!; “ second interval’
LO CA TE 2, 1
PR IN T “N o. o f samples to be taken” ; endtim % * cycle!;
93
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LOCATE 3, 1
PR IN T ‘T otal time taken
tottim% ; “ seconds” ;
LOCATE 5, 1
PR IN T “W avelength[A] C urrent[am ps] Sam ple No. Cycle/Sec. Time rem aining[sec.]
start! = TIM ER
DO
L O O P W H ILE i% < endtim % + 1
finish! = TIM ER
TIM ER OFF
PR IN T ‘T otal Execution tim e=” ; finish! - start!
PR IN T # 1 ,E N $
PR IN T #2, EN$
O PE N path$ + “\X ” + filenmS + “ .m” F O R O U TPU T AS #3 ’store m atlab E X E file
PR IN T #3, “clear all;clc;”
PR IN T #3, “W ” + filenmS
PR IN T #3, “D ” + filenmS
PR IN T #3, “plot(waveln,abs(value),’g ’);pause(3);sm 5;xlabel(‘W avelength A’);ylab e l(‘Current am ps’);”
PR IN T #3, “title(‘Em ission Spectrum ’);grid;”
PR IN T “Execute “ ; pathS + “\X ” + filenmS + “ .m” ; “ in matlab to view p lo t”
CLO SE
GOTO 200
END
disp:
FO R cyc% = 1 TO cycle!
CALL FDLEWRT
count! = count! + 1
N EX T cyc%
i% = i% + L
RETURN
200 CO LO R w hite, red
PRIN T “Sw itch off the Spectrom eter drive” ;
CO LOR white, black
PRIN T “ Press any key to end.”;
W H ILE INKEYS = ““
HTONE = 2000: LTONE = 550: DELA Y = 500
FOR count = HTO N E TO LTONE ST E P -10
SOUND count, DELAY / count
N EX T count
HTONE = 780: R A N G E = 650
FO R count = RA N G E TO -RA N G E ST EP -4
94
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
SO U N D H TONE - A B S(count), .3
count = count - 2 / R A N G E
N EX T count
W END
EN D IF
END
SUB FILEW RT
SH A RED prev, pico% , cyc% , i%, stain!, resol!, tottim % , sec% , tem p$, buff$, count!
C A L L IBRD (pico% , buff$)
tem p$ = M ID$(buff$, 5, 40)
y = VAL(RTRIM $(temp$))
IF (y > .00001) TH EN
y = prev
EN D IF
prev = y
stain! = stain! + resol!
W RITE #1, 2 * stain!
W RITE #2, y
LOCATE 7, 1
PR IN T 2 * stain!, y, count!, cyc% , tottim % - (sec% * i%)
EN D SUB
SUB R eportError (fd% , errmsgS) STATIC
PRIN T “E rror = “ , EBERR%; errm sgS
EF (fd% o -1) TH E N
PR IN T (“Cleanup: taking board off-line”)
C A LL EBONL(fd%, 0)
END IF
STOP
‘ A bort program
END SUB
95
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
REFERENCES
96
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
REFERENCES
[1] J. A sm ussen, J. M ossbrucker, S. K hatam i, W. S. H uang, B. W right and V. Ayres, ‘T h e
effect o f nitrogen on the grow th, m orphology, and crystalline quality o f M PACVD
diam ond films,” Subm itted to Diamond and related materials.
[2] D . Z hou, D. M. G ruen, L. C. Q in, T. G . M cC auley and A. R. K rauss, “C ontrol o f dia­
m o n d film m icrostructure by A r additions to C H 4 /H 2 m icrow ave plasm as,” J. Appl.
Phys., 84, No. 4, 1998
[3] A. K. Srivastava, Properties o f E lectron C yclotron R esonance Plasm a Sources. PhJD.
D issertation, M ichigan State U niversity, 1995.
[4] A. R. Striganov and N. S. Sventitskii, Tables o f Spectral Lines o f N eutrals and Ionized
A to m s. (IFI/Plenum , New York, 1968).
[5] H. N. Chu, E. A. Den Hartog, A. R. Lefkow, J. Jacobs, L. W. A nderson, M. G. Lefally
and J. E. Lawler, “M easurem ents o f the gas tem perature in a C H 4 -H 2 discharge during
the grow th of diam ond,” Phys. Rev. A, 44, 6 , 1991.
[6 ] A. N . G oyette, J. R. Peck, Y. M atsuda, L. W. A nderson and J. E. Law ler, “Experim en­
tal com parison o f rotational and gas kinetic tem peratures in N 2 and H e-N 2 dis­
charges,” J. Phys. D: Appl. Phys, 31, 1556, (1998).
[7] A. T h o m e, Spectrophvsics (C hapm an and Hall, New York, 1988)
[8 ] A. G icquel, K. H assouni, Y. B reton, M . C henevier and J. C. C ubertafon, “ Gas tem per­
ature m easurem ents by laser spectroscopic techniques and by optical em ission spec­
troscopy,” Diamond and related materials, 5, 366, (1996).
[9] L. W olniewicz and K. Dressier, ‘T h e EF and GK
+ States o f H ydrogen- A diabatic
C alculation of Vibronic States in H 2, HD, and D 2,” J. Mol. Spectrosc., 67,416, (1977).
[10] K. D ressier, R. Gallusser, P. Q uadrelli and L. W olniewicz, ‘T h e EF and GK ^ g +
States o f Hydrogen- Calculation o f N onadiabatic C oupling,” J. Mol. Spectrosc., 75,
205, (1979).
[11] I. Kovacs, Rotational Structure in the Spectra o f D iatom ic m olecules (A m erican
E lsevier Publishing Company, N ew York, 1969)
[12] G . H . Dieke, ‘T h e M olecular Spectrum o f H ydrogen and its Isotopes,” J.Mol. Spec­
trosc., 2, 494, (1958).
97
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
[13] G. Herzberg, M olecular Spectra and M olecular Structure vol 1. (Van Nostrand, New
York, 1950).
[14] R. Sam lenski, C. Haug a n d R. Brenn, “Incorporation o f nitrogen in chemical vapor
deposited diamond,” Appl. Phys. Lett., 67, No. 19, 2798, 1995.
[15] W. M uller-Sebert, E. W om er, F. Fuchs, C. W ild and P. K iodl, “Nitrogen induced
increase o f growth rate in chem ical vapor deposition o f diam ond,” Appl. Phys. Lett.,
6 8 , No. 6 , 759, 1996.
[16] T. Hong, S. Chen, Y. C hiou and C. Chen, “Optical em ission spectroscopy studies o f
the effects o f nitrogen addition on diam ond synthesis in a C H 4 -CO 2 gas mixture,”
Appl. Phys. Lett., 61, No. 15, 2149. 1995.
[17] W. Zhu, A. Inspektor, A. R. Badzian, T. M ckenna and R. M essier, “Effects o f noblegases on diam ond deposition from methane-hydro gen m icrowave plasmas,” J. Appl.
Phys., 6 8 , 1489, 1990.
[18] K. J. Clay, S. P. Speakm an, G. A. J. A m aratunga and S. R. P. Silva, “Characterization
o f a-C :H :N deposition from C H 4 /N 2 rf plasmas using optical em ission spectroscopy,”
/. Appl. Phys., 79, No. 9, 7227, 1996.
98
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
IMAGE EVALUATION
TEST TARGET (Q A - 3 )
j f i a
1 .0
U
i.
l.l
gia.”. lU!- 1
1 .2 5
£
£ L£
Li
L_
L_U.i_
m
[I 2.2
HI 2 .0
1.8
1 .4
1.6
150mm
IIV U G E .Inc
1653 East Main Street
Rochester, NY 14609 USA
Phone: 716/482-0300
Fax: 716/288-5989
O 1 9 9 3 . A p p lie d I m a g e . In c .. All R ig h t s R e s e r v e d
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Документ
Категория
Без категории
Просмотров
0
Размер файла
3 300 Кб
Теги
sdewsdweddes
1/--страниц
Пожаловаться на содержимое документа